{"id":46266,"date":"2023-11-04T22:51:10","date_gmt":"2023-11-04T14:51:10","guid":{"rendered":"https:\/\/swarma.org\/?p=46266"},"modified":"2023-11-04T22:51:10","modified_gmt":"2023-11-04T14:51:10","slug":"20%e5%a4%9a%e4%bd%8d%e5%ad%a6%e8%80%85%e5%85%b1%e8%af%9d%e5%9b%a0%e6%9e%9c%e7%a7%91%e5%ad%a6%e6%96%b0%e8%bf%9b%e5%b1%95-2023%e6%b3%9b%e5%a4%aa%e5%b9%b3%e6%b4%8b%e5%9b%a0%e6%9e%9c","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=46266","title":{"rendered":"20\u591a\u4f4d\u5b66\u8005\u5171\u8bdd\u56e0\u679c\u79d1\u5b66\u65b0\u8fdb\u5c55\u2014\u20142023\u6cdb\u592a\u5e73\u6d0b\u56e0\u679c\u63a8\u65ad\u5927\u4f1a\u56de\u987e\u603b\u7ed3"},"content":{"rendered":"
\n
\n
\n
\n
<\/div>\n<\/section>\n
\n
\n
\n
\n
\n
\n

<\/section>\n<\/section>\n
\n
\n

\u5bfc\u8bed<\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\u56e0\u679c\u63a8\u65ad\u7684\u76ee\u6807\u662f\u5c06\u5916\u90e8\u77e5\u8bc6\u4e0e\u5b66\u4e60\u8bbe\u8ba1\u7ed3\u5408\u8d77\u6765\uff0c\u5f97\u5230\u53d8\u91cf\u4e4b\u95f4\u7684\u56e0\u679c\u5173\u7cfb\u3002\u5c24\u5176\u662f\u5728\u4eba\u5de5\u667a\u80fd\u6280\u672f\u906d\u9047\u74f6\u9888\u7684\u73b0\u5728\uff0c\u56e0\u679c\u63a8\u7406\u7684\u65b9\u6cd5\u4e5f\u8d8a\u6765\u8d8a\u53d7\u5230\u4e0d\u540c\u9886\u57df\u7684\u91cd\u89c6\uff0c\u5b83\u5728\u4eba\u5de5\u667a\u80fd\u3001\u7edf\u8ba1\u5b66\u3001\u751f\u7269\u7edf\u8ba1\u5b66\u3001\u751f\u7269\u533b\u5b66\u3001\u8ba1\u7b97\u673a\u79d1\u5b66\u3001\u7ecf\u6d4e\u5b66\u3001\u6d41\u884c\u75c5\u5b66\u548c\u5404\u79cd\u793e\u4f1a\u79d1\u5b66\u7b49\u9886\u57df\u83b7\u5f97\u4e86\u5e7f\u6cdb\u7684\u5e94\u7528\u3002<\/span><\/strong><\/section>\n

<\/span><\/strong><\/section>\n
\u4e3a\u4e86\u66f4\u591a\u5730\u63a8\u52a8\u56e0\u679c\u79d1\u5b66\u5b66\u79d1\u7684\u53d1\u5c55\uff0c\u805a\u96c6\u56fd\u5185\u5916\u56e0\u679c\u79d1\u5b66\u7684\u4e00\u7ebf\u79d1\u7814\u5de5\u4f5c\u8005\uff0c\u5171\u540c\u8ba8\u8bba\u56e0\u679c\u79d1\u5b66\u7684\u6700\u65b0\u8fdb\u5c55\uff0c\u5317\u4eac\u5927\u5b66\u8bb2\u5e2d\u6559\u6388\u3001\u5317\u4eac\u5927\u5b66\u516c\u5171\u536b\u751f\u5b66\u9662\u751f\u7269\u7edf\u8ba1\u7cfb\u7cfb\u4e3b\u4efb\u3001\u5317\u4eac\u5927\u5b66\u5317\u4eac\u56fd\u9645\u6570\u5b66\u7814\u7a76\u4e2d\u5fc3\u751f\u7269\u7edf\u8ba1\u548c\u4fe1\u606f\u7814\u7a76\u5ba4\u4e3b\u4efb\u5468\u6653\u534e\u7b49\u53d1\u8d77\u4e86\u6cdb\u592a\u5e73\u6d0b\u56e0\u679c\u63a8\u65ad\u5927\u4f1a\u30022023\u6cdb\u592a\u5e73\u6d0b\u56e0\u679c\u63a8\u65ad<\/a>\u5927\u4f1a\u4e8e\u4eca\u5e749\u6708\u4e3e\u884c\uff0c\u96c6\u667a\u4ff1\u4e50\u90e8\u56e0\u679c\u79d1\u5b66\u793e\u533a\u6210\u5458\u6574\u7406\u4e86\u6b64\u6b21\u5927\u4f1a\u4e2d20\u591a\u4f4d\u5b66\u8005\u5173\u4e8e\u56e0\u679c\u79d1\u5b66\u4e0d\u540c\u65b9\u5411\u7684\u62a5\u544a\uff0c\u5206\u4eab\u7ed9\u5927\u5bb6\u3002<\/span><\/strong><\/section>\n<\/section>\n
\n

<\/section>\n<\/section>\n
\n
\u7814\u7a76\u9886\u57df\uff1a\u56e0\u679c\u79d1\u5b66\uff0c\u56e0\u679c\u53d1\u73b0\uff0c\u56e0\u679c\u8868\u5f81\u5b66\u4e60<\/span><\/strong><\/strong><\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/section>\n<\/section>\n
\n
<\/strong><\/section>\n<\/section>\n
\n
<\/strong><\/strong><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
<\/section>\n<\/section>\n

<\/section>\n

<\/section>\n


<\/h1>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

\u968f\u673a\u8bd5\u9a8c\u4e2d\u610f\u56fe\u6cbb\u7597\u6bd4\u8f83\u7684\u56e0\u679c\u5173\u7cfb<\/span><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n

\n
\u4e3b\u9898\uff1aCausality of Intention-to-Treat Comparisons in Randomized Trials<\/span><\/section>\n
\u4e3b\u8bb2\u4eba\uff1aRoss L. Prentice<\/span><\/section>\n
\u7f8e\u56fd Fred Hutchinson \u764c\u75c7\u4e2d\u5fc3\u548c\u534e\u76db\u987f\u5927\u5b66\uff0c\u524d\u9ad8\u7ea7\u526f\u603b\u88c1\u517cHutch\u516c\u5171\u536b\u751f\u79d1\u5b66\u90e8\u4e3b\u4efb\u3002Ross Prentice\u00eds\u535a\u58eb\u662f\u4e00\u4f4d\u751f\u7269\u7edf\u8ba1\u5b66\u5bb6\uff0c\u4ed6\u5f00\u53d1\u4e86\u7528\u4e8e\u4e34\u5e8a\u548c\u4eba\u7fa4\u7814\u7a76\u8bbe\u8ba1\u548c\u5206\u6790\u7684\u65b9\u6cd5\u3002\u4ed6\u5bf9\u9884\u9632\u6162\u6027\u75c5\u7684\u7814\u7a76\u7279\u522b\u611f\u5174\u8da3\uff0c\u91cd\u70b9\u662f\u996e\u98df\u548c\u6fc0\u7d20\u7684\u53d8\u5316\uff0c\u8fd9\u53ef\u80fd\u4f1a\u964d\u4f4e\u764c\u75c7\u548c\u5fc3\u8840\u7ba1\u75be\u75c5\u7684\u98ce\u9669\u3002<\/span><\/section>\n
\u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
fredhutch.org\/en\/faculty\/lab-directory\/prentice-ross.html<\/span><\/section>\n<\/blockquote>\n


<\/h2>\n

\u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
Presentation based on Prentice RL and Aragaki AK. Intention-to-treat comparisons in randomized trials. Statistical Science 2022;37:380-393.<\/span><\/section>\n


<\/h2>\n

\u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

\u6458\u8981<\/span><\/strong><\/h3>\n

\u610f\u5411\u6cbb\u7597\uff08ITT\uff09\u6bd4\u8f83\u5728\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u7684\u62a5\u544a\u4e2d\u5177\u6709\u6838\u5fc3\u5730\u4f4d\uff0c\u5c3d\u7ba1\u901a\u5e38\u8fd8\u6709\u5176\u4ed6\u5229\u76ca\u5206\u6790\uff0c\u4f8b\u5982\u5bf9\u4e0d\u9075\u5b88\u505a\u51fa\u8c03\u6574\u7684\u5206\u6790\u3002\u5728\u6211\u4eec\u5bf9\u5987\u5973\u5065\u5eb7\u5021\u8bae\uff08WHI\uff09\u7ed3\u679c\u7684ITT\u62a5\u544a\u4e2d\u968f\u673a\u8bd5\u9a8c\uff0c\u6211\u4eec\u4e3b\u8981\u4f9d\u9760\u9ad8\u5ea6\u7075\u6d3b\u7684\u98ce\u9669\u6bd4<\/span><\/p>\n

\uff08Cox\uff09\u56de\u5f52\u65b9\u6cd5\u3002\u7136\u800c\uff0c\u8fd9\u4e9b\u65b9\u6cd5\uff0c\u7279\u522b\u662f\u6bd4\u4f8b\u5371\u9669\u7279\u4f8b\uff0c\u88ab\u6279\u8bc4\u4e3a\u96be\u4ee5\u89e3\u91ca\uff0c\u7ecf\u5e38\u8fc7\u4e8e\u7b80\u5355\u5316\uff0c\u5e76\u4e14\u4e0e\u73b0\u4ee3\u65b9\u6cd5\u4e0d\u4e00\u81f4\u4f7f\u7528\u6f5c\u5728\u7ed3\u679c\u7684\u56e0\u679c\u5173\u7cfb\u7406\u8bba\u3002\u5728\u8fd9\u91cc\uff0c\u6211\u4eec\u8ba8\u8bba\u8fd9\u4e9b\u4e3b\u9898\u548c\u6269\u5927\u4e86\u6211\u4eec\u5728WHI\u8bd5\u9a8c\u4e2dITT\u6bd4\u8f83\u7684\u5371\u9669\u7387\u65b9\u6cd5\u7684\u4f7f\u7528\u8303\u56f4\u3002<\/span><\/p>\n

\u521b\u65b0<\/span><\/strong><\/h3>\n
\u6587\u7ae0\u9996\u6b21\u63a2\u7d22\u4e86\u5728\u968f\u673a\u8bd5\u9a8c\u62a5\u544a\u4e2d\u4f7f\u7528\u5371\u9669\u6bd4\u65b9\u6cd5\u7684\u5b9e\u8df5\u3002\u5177\u4f53\u6765\u8bf4\uff0c\u5b83\u5173\u6ce8\u4e86\u5987\u5973\u5065\u5eb7\u5021\u8bae\u4e2d\u8fdb\u884c\u7684\u968f\u673a\u5bf9\u7167\u6fc0\u7d20\u7597\u6cd5\u8bd5\u9a8c\u3002\u4f5c\u8005\u6307\u51fa\uff0c\u5148\u524d\u7684\u7814\u7a76\u4e2d\u5e38\u5e38\u5b58\u5728\u4f30\u8ba1\u7684\u5371\u9669\u6bd4\uff08HR\uff09\u5728\u7279\u5b9a\u65f6\u95f4\u70b9 t>0 \u5b58\u5728\u5185\u5728\u7684\u201c\u9009\u62e9\u504f\u5dee\u201d\u3002\u6b64\u5916\uff0c\u4e00\u4e9b\u5b66\u8005\u8ba4\u4e3a\u5371\u9669\u6bd4\uff08HR\uff09\u672c\u8eab\u5e76\u4e0d\u9002\u5408\u8fdb\u884c\u56e0\u679c\u89e3\u91ca\uff0c\u5c24\u5176\u662f\u5728\u5b58\u5728\u672a\u5efa\u6a21\u7684\u5f02\u8d28\u6027\u7684\u60c5\u51b5\u4e0b\u3002\u56e0\u6b64\uff0c\u4ed6\u4eec\u63d0\u51fa\u4e86\u4f7f\u7528\u66f4\u7b26\u5408\u56e0\u679c\u63a8\u7406\u7684\u5efa\u6a21\u6846\u67b6\u6765\u8fdb\u884c\u4f30\u8ba1\u7684\u89c2\u70b9\u3002\u6587\u7ae0\u7684\u521b\u65b0\u70b9\u5728\u4e8e\uff0c\u4f5c\u8005\u4e3b\u5f20\u6574\u4e2a\u5371\u9669\u6bd4\u51fd\u6570\u4e0e\u968f\u673a\u8bd5\u9a8c\u4e2d\u7684\u6cbb\u7597\u56e0\u679c\u76f8\u5173\uff0c\u5f3a\u8c03\u4e86\u5bf9\u5e73\u5747\u5371\u9669\u6bd4\u4f30\u8ba1\u7684\u91cd\u8981\u6027\uff0c\u5e76\u5c06\u8fd9\u4e00\u65b9\u6cd5\u5e94\u7528\u4e8e\u5987\u5973\u5065\u5eb7\u5021\u8bae\u7684\u6fc0\u7d20\u7597\u6cd5\u8bd5\u9a8c\u4e2d\u3002\u8fd9\u4e9b\u521b\u65b0\u70b9\u4e3a\u56e0\u679c\u63a8\u65ad\u63d0\u4f9b\u4e86\u65b0\u7684\u89c6\u89d2\u548c\u65b9\u6cd5\uff0c\u5c24\u5176\u5173\u6ce8\u4e86\u5371\u9669\u6bd4\u51fd\u6570\u5728\u968f\u673a\u8bd5\u9a8c\u4e2d\u7684\u56e0\u679c\u5173\u7cfb\u3002<\/span><\/section>\n

<\/span><\/section>\n
\u6574\u7406\u8005\uff1a\u82cf\u5a49\u5a77<\/span><\/section>\n

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

\u56e0\u679c\u63a8\u65ad\u4e0e\u53cd\u4e8b\u5b9e\u9884\u6d4b\u6cbb\u7597\u6548\u679c\uff1a<\/strong><\/p>\n

\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u548c\u89c2\u5bdf\u6027\u7814\u7a76\u7684\u5efa\u6a21<\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n

\n
\u4e3b\u9898: Causal inference and counterfactual prediction of treatment benefit: modeling in randomized controlled trials and observational studies<\/span><\/section>\n
\u4e3b\u8bb2\u4eba\uff1aEwout Steyerberg<\/span><\/section>\n
Ewout Steyerberg \u662f\u8377\u5170\u83b1\u987f\u5927\u5b66\u533b\u5b66\u4e2d\u5fc3\u548cErasmus MC\uff08\u8377\u5170\u9e7f\u7279\u4e39\uff09\u7684\u4e34\u5e8a\u751f\u7269\u7edf\u8ba1\u5b66\u548c\u533b\u5b66\u51b3\u7b56\u5236\u5b9a\u6559\u6388\u3002\u81ea2017\u5e74\u4ee5\u6765\uff0c\u4ed6\u4e00\u76f4\u62c5\u4efb\u751f\u7269\u533b\u5b66\u6570\u636e\u79d1\u5b66\u7cfb\u7684\u4e3b\u5e2d\u3002\u4ed6\u662f\u8377\u5170\u7687\u5bb6\u827a\u672f\u4e0e\u79d1\u5b66\u5b66\u4f1a\uff08KNAW\uff09\u7684\u4f1a\u5458\u3002\u4ed6\u66fe\u5728\u675c\u514b\u5927\u5b66\uff08\u5317\u5361\u7f57\u6765\u7eb3\u5dde\u8fbe\u52d2\u59c6\uff1a1996\u5e74\uff09\u548c\u54c8\u4f5b\u5927\u5b66\uff08\u9a6c\u8428\u8bf8\u585e\u5dde\u6ce2\u58eb\u987f\uff1a2003\u5e74\u548c2005\u5e74\uff09\u5ea6\u8fc7\u4e86\u5b66\u672f\u751f\u6daf\u3002<\/span><\/section>\n<\/blockquote>\n


<\/h2>\n

\u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
    \n
  1. \n

    GUSTO-I<\/span><\/p>\n<\/li>\n

      \n
    1. \n

      RCT in NEJM 1993<\/span><\/p>\n<\/li>\n

    2. \n

      Risk-based analyses Califf AHJ 1997<\/span><\/p>\n<\/li>\n

    3. \n

      Covariate adjustment Steyerberg 2000<\/span><\/p>\n<\/li>\n

    4. \n

      https:\/\/www.fharrel.com\/post\/path\/<\/span><\/p>\n<\/li>\n<\/ol>\n

    5. \n

      PATH Statement<\/span><\/p>\n<\/li>\n

        \n
      1. \n

        Ann Int Med 2020<\/span><\/p>\n<\/li>\n<\/ol>\n

      2. \n

        Risks of traditional subgroup analyses<\/span><\/p>\n<\/li>\n

          \n
        1. \n
          Assmann Lancet 2000; Rothwell 2005; Hernandez 2006; Wallach BMJ 2016 & JAMA Int Med 2017<\/span><\/section>\n<\/li>\n<\/ol>\n<\/ol>\n


          <\/h2>\n

          \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n
            \n
          1. \n

            \u56e0\u679c\u63a8\u65ad\u7684\u57fa\u672c\u6982\u5ff5\uff1a\u6f14\u8bb2\u8005\u89e3\u91ca\u4e86\u56e0\u679c\u63a8\u65ad\u7684\u6982\u5ff5\uff0c\u5e76\u5f3a\u8c03\u4e86\u5b83\u5728\u7814\u7a76\u4e2d\u7684\u91cd\u8981\u6027\u3002\u4ed6\u6307\u51fa\uff0c\u56e0\u679c\u63a8\u65ad\u662f\u786e\u5b9a\u72ec\u7acb\u5b9e\u9645\u6548\u5e94\u7684\u8fc7\u7a0b\uff0c\u5373\u786e\u5b9a\u4e00\u4e2a\u7279\u5b9a\u73b0\u8c61\u7684\u56e0\u679c\u5173\u7cfb\u3002\u4ed6\u63d0\u5230\u4e86\u76f8\u5173\u6027\u548c\u56e0\u679c\u6027\u4e4b\u95f4\u7684\u533a\u522b\uff0c\u5e76\u8b66\u544a\u4eba\u4eec\u4e0d\u8981\u9519\u8bef\u5730\u8ba4\u4e3a\u76f8\u5173\u6027\u5c31\u610f\u5473\u7740\u56e0\u679c\u6027\u3002\u6f14\u8bb2\u8005\u8fd8\u63d0\u5230\u4e86\u56e0\u679c\u63a8\u65ad\u5728\u533b\u5b66\u548c\u5176\u4ed6\u9886\u57df\u4e2d\u7684\u5e7f\u6cdb\u5e94\u7528\uff0c\u4ee5\u53ca\u5b83\u5728\u51b3\u7b56\u5236\u5b9a\u548c\u653f\u7b56\u5236\u5b9a\u4e2d\u7684\u91cd\u8981\u6027\u3002<\/span><\/p>\n<\/li>\n

          2. \n

            \u89c2\u5bdf\u6027\u6570\u636e\u548c\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\uff1a\u6f14\u8bb2\u8005\u8be6\u7ec6\u8ba8\u8bba\u4e86\u89c2\u5bdf\u6027\u6570\u636e\u548c\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u7684\u4e0d\u540c\u4f5c\u7528\u3002\u4ed6\u89e3\u91ca\u4e86\u89c2\u5bdf\u6027\u6570\u636e\u5206\u6790\u9700\u8981\u8003\u8651\u5404\u79cd\u5047\u8bbe\u548c\u8fd1\u4f3c\uff0c\u800c\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u5219\u63d0\u4f9b\u4e86\u66f4\u5f3a\u7684\u8bc1\u636e\u652f\u6301\u56e0\u679c\u5173\u7cfb\u3002\u4ed6\u63d0\u5230\u4e86\u610f\u5411\u6cbb\u7597\u5206\u6790\u4f5c\u4e3a\u4e00\u79cd\u5728\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u4e2d\u5e38\u7528\u7684\u65b9\u6cd5\uff0c\u540c\u65f6\u4e5f\u6307\u51fa\u4e86\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\u53ef\u80fd\u9047\u5230\u7684\u6311\u6218\u548c\u9650\u5236\u3002<\/span><\/p>\n<\/li>\n

          3. \n

            \u9884\u6d4b\u56e0\u7d20\u548c\u9884\u540e\u56e0\u7d20\uff1a\u6f14\u8bb2\u8005\u8be6\u7ec6\u4ecb\u7ecd\u4e86\u9884\u6d4b\u56e0\u7d20\u548c\u9884\u540e\u56e0\u7d20\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u7684\u4f5c\u7528\u548c\u5dee\u5f02\u3002\u4ed6\u89e3\u91ca\u4e86\u9884\u6d4b\u56e0\u7d20\u662f\u57fa\u4e8e\u7279\u5f81\u548c\u6cbb\u7597\u6548\u679c\u7684\u4e3b\u8981\u6548\u5e94\uff0c\u7528\u4e8e\u9884\u6d4b\u60a3\u8005\u7684\u7ed3\u679c\u3002\u800c\u9884\u540e\u56e0\u7d20\u5219\u6d89\u53ca\u6cbb\u7597\u6548\u679c\u548c\u9884\u540e\u56e0\u7d20\u4e4b\u95f4\u7684\u4ea4\u4e92\u4f5c\u7528\uff0c\u5b83\u4eec\u80fd\u591f\u63d0\u4f9b\u66f4\u4e3a\u7cbe\u786e\u7684\u4e2a\u4f53\u5316\u6cbb\u7597\u4fe1\u606f\u3002\u6f14\u8bb2\u8005\u5f3a\u8c03\u4e86\u9884\u540e\u56e0\u7d20\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u7684\u91cd\u8981\u6027\uff0c\u5e76\u6307\u51fa\u5229\u7528\u9884\u540e\u56e0\u7d20\u8fdb\u884c\u5206\u6790\u53ef\u4ee5\u66f4\u597d\u5730\u5e94\u5bf9\u56e0\u679c\u63a8\u65ad\u7684\u9700\u6c42\uff0c\u800c\u8fc7\u5ea6\u5173\u6ce8\u9884\u6d4b\u56e0\u7d20\u53ef\u80fd\u5bfc\u81f4\u865a\u5047\u7ed3\u679c\u548c\u8bef\u5bfc\u3002<\/span><\/p>\n<\/li>\n

          4. \n
            \u4e2a\u4f53\u5316\u62a4\u7406\u548c\u7cbe\u51c6\u533b\u5b66\uff1a\u6f14\u8bb2\u8005\u63a2\u8ba8\u4e86\u4e2a\u4f53\u5316\u62a4\u7406\u548c\u7cbe\u51c6\u533b\u5b66\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u7684\u5e94\u7528\u3002\u4ed6\u63d0\u5230\u4e86\u7cbe\u51c6\u533b\u5b66\u4e2d\u5173\u4e8e\u4e2a\u4f53\u7279\u5f81\u548c\u6cbb\u7597\u6548\u679c\u7684\u7814\u7a76\uff0c\u8fd9\u4e9b\u7814\u7a76\u65e8\u5728\u6839\u636e\u60a3\u8005\u7684\u4e2a\u4f53\u7279\u5f81\u548c\u9884\u540e\u56e0\u7d20\u6765\u5236\u5b9a\u4e2a\u6027\u5316\u7684\u6cbb\u7597\u65b9\u6848\u3002\u7136\u800c\uff0c\u6f14\u8bb2\u8005\u4e5f\u6307\u51fa\u4e86\u5c06\u56e0\u679c\u63a8\u65ad\u5e94\u7528\u4e8e\u4e2a\u4f53\u5316\u62a4\u7406\u7684\u6311\u6218\uff0c\u5373\u5982\u4f55\u5c06\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\u4e2d\u7684\u7ed3\u679c\u8f6c\u5316\u4e3a\u9002\u7528\u4e8e\u4e2a\u522b\u60a3\u8005\u7684\u6cbb\u7597\u7b56\u7565\u3002\u4ed6\u5f3a\u8c03\u4e86\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u66f4\u91cd\u8981\u7684\u662f\u5229\u7528\u9884\u540e\u56e0\u7d20\u800c\u975e\u8fc7\u5ea6\u8ffd\u6c42\u9884\u6d4b\u56e0\u7d20\uff0c\u4ee5\u5b9e\u73b0\u66f4\u9ad8\u6548\u7684\u4e2a\u4f53\u5316\u62a4\u7406\u548c\u51b3\u7b56\u5236\u5b9a\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

            <\/section>\n
            \u6574\u7406\u8005\uff1a\u8d21\u7545<\/span><\/section>\n


            <\/span><\/p>\n


            <\/span><\/section>\n

            \n
            \n
            \n
            \n
            \n

            <\/section>\n

            <\/section>\n
            \n
            <\/svg><\/section>\n<\/section>\n<\/section>\n
            \n

            \u9152\u7cbe\u4e0e\u5fc3\u8840\u7ba1\u75be\u75c5\uff1a<\/strong><\/p>\n

            \u9002\u5ea6\u996e\u9152\u771f\u7684\u5bf9\u5fc3\u8840\u7ba1\u75be\u75c5\u6709\u76ca\u5417?<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

            \n

            <\/section>\n

            <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


            <\/p>\n

            \n
            \u4e3b\u9898\uff1aAlcohol and cardiovascular disease: is moderate drinking really beneficial for cardiovascular disease?<\/span><\/section>\n
            \u4e3b\u8bb2\u4eba\uff1aDerrick Bennett \uff0c\u725b\u6d25\u5927\u5b66 Nuffield Department of Population Health \u526f\u6559\u6388<\/span><\/section>\n
            \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/www.ndph.ox.ac.uk\/team\/derrick-bennett<\/span><\/section>\n<\/blockquote>\n


            <\/h2>\n

            \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

            Ferri\u00e8res J. The French paradox: lessons for other countries[J]. Heart, 2004, 90(1): 107-111.<\/span><\/p>\n

            Emberson J R, Bennett D A. Effect of alcohol on risk of coronary heart disease and stroke: causality, bias, or a bit of both?[J]. Vascular health and risk management, 2006, 2(3): 239-249.<\/span><\/p>\n

            Wood A M, Kaptoge S, Butterworth A S, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies[J]. The Lancet, 2018, 391(10129): 1513-1523.<\/span><\/p>\n

            Millwood I Y, Walters R G, Mei X W, et al. Conventional and genetic evidence on alcohol and vascular disease aetiology: a prospective study of 500 000 men and women in China[J]. The Lancet, 2019, 393(10183): 1831-1842.<\/span><\/section>\n


            <\/h2>\n

            \u5185\u5bb9<\/span><\/strong><\/h2>\n
            \u62a5\u544a\u8ba8\u8bba\u4e86\u9152\u7cbe\u4e0e\u516c\u5171\u536b\u751f\u95ee\u9898\u4e4b\u95f4\u7684\u5173\u7cfb\uff0c\u4e3b\u8981\u96c6\u4e2d\u5728\u9152\u7cbe\u6444\u5165\u4e0e\u5065\u5eb7\u6548\u5e94\u4e4b\u95f4\u7684\u8054\u7cfb\u3002\u5176\u4e2d\uff0c\u63d0\u5230\u4e86\u4e00\u4e9b\u65e9\u671f\u7814\u7a76\uff0c\u5176\u4e2d\u53d1\u73b0\u4e86\u4e00\u79cd\u79f0\u4e3a\u201c\u6cd5\u56fd\u6096\u8bba\u201d\u7684\u73b0\u8c61\uff0c\u5373\u5c3d\u7ba1\u6cd5\u56fd\u4eba\u7684\u80c6\u56fa\u9187\u6c34\u5e73\u5f88\u9ad8\uff0c\u4f46\u4ed6\u4eec\u7684\u5fc3\u8840\u7ba1\u75be\u75c5\u6b7b\u4ea1\u7387\u8f83\u4f4e\uff0c\u8fd9\u53ef\u80fd\u4e0e\u4ed6\u4eec\u5728\u996e\u98df\u4e2d\u7684\u9ad8\u8461\u8404\u9152\u6d88\u8d39\u6709\u5173\u3002\u6b64\u5916\uff0c\u8fd8\u8ba8\u8bba\u4e86\u5438\u70df\u4e0e\u80ba\u764c\u7684\u7c7b\u4f3c\u5173\u8054\uff0c\u5e76\u63d0\u5230\u4e86\u996e\u9152\u6a21\u5f0f\u5728\u4e0d\u540c\u56fd\u5bb6\u4e4b\u95f4\u7684\u5dee\u5f02\u3002\u7136\u540e\uff0c\u4ecb\u7ecd\u4e86\u57fa\u56e0\u7814\u7a76\uff0c\u5176\u4e2d\u6d89\u53ca\u4e24\u79cd\u5f71\u54cd\u9152\u7cbe\u4ee3\u8c22\u7684\u57fa\u56e0\u53d8\u5f02\uff0c\u4ee5\u53ca\u5982\u4f55\u4f7f\u7528\u8fd9\u4e9b\u57fa\u56e0\u6765\u7814\u7a76\u9152\u7cbe\u6444\u5165\u4e0e\u5065\u5eb7\u4e4b\u95f4\u7684\u5173\u7cfb\u3002<\/span><\/section>\n

            <\/section>\n
            \u6574\u7406\u8005\uff1a\u738b\u6587\u793c<\/span><\/section>\n

            <\/section>\n


            <\/span><\/p>\n

            \n
            \n
            \n
            \n
            \n

            <\/section>\n

            <\/section>\n
            \n
            <\/svg><\/section>\n<\/section>\n<\/section>\n
            \n

            \u4f7f\u7528TMLE\u5bf9\u91cd\u75c7\u76d1\u62a4\u4e2d\u7684\u56db\u4e2a<\/strong><\/p>\n

            \u5927\u578b\u968f\u673a\u5bf9\u7167\u5b9e\u9a8c\u518d\u5206\u6790\uff1b\u662f\u5426\u6ee1\u8db3\u7406\u8bba\u627f\u8bfa<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

            \n

            <\/section>\n

            <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


            <\/p>\n

            \n
            \u4e3b\u9898\uff1aReanalyzing a four large RCTs within intensive care using TMLE; does it deliver on the theoretical promises<\/span><\/section>\n
            \u4e3b\u8bb2\u4eba\uff1aTheis Lange<\/span><\/section>\n
            Theis Lange \u662f\u54e5\u672c\u54c8\u6839\u5927\u5b66\u751f\u7269\u7edf\u8ba1\u5b66\u6559\u6388\uff0c\u83b7\u5f97\u54e5\u672c\u54c8\u6839\u5927\u5b66\u6570\u7406\u7edf\u8ba1\u535a\u58eb\u5b66\u4f4d\u3002\u7814\u7a76\u5174\u8da3\u5305\u62ec\u56e0\u679c\u63a8\u65ad\u3001\u4e34\u5e8a\u8bd5\u9a8c\u7edf\u8ba1\u5206\u6790\u3001\u975e\u7ebf\u6027\u52a8\u6001\u6a21\u578b\uff1b\u76ee\u524d\u7814\u7a76\u5305\u62ec\u5206\u6790\u764c\u75c7\u4e2d\u9020\u6210\u793e\u4f1a\u4e0d\u5e73\u7b49\u7684\u56e0\u679c\u9014\u5f84\uff0c\u6bd4\u8f83\u751f\u5b58\u5206\u6790\u80cc\u666f\u548c\u57fa\u56e0-\u75be\u75c5\u5173\u8054\u7814\u7a76\u4e2d\u7684\u4e0d\u540c\u8c03\u8282\u63aa\u65bd\u7b49\u3002<\/span><\/section>\n
            \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/dsin.ku.dk\/key_researchers\/lange\/<\/span><\/section>\n<\/blockquote>\n


            <\/h2>\n

            \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

            \u6458\u8981<\/span><\/strong><\/h3>\n

            \u8fd9\u6bb5\u62a5\u544a\u4e2d\u4ecb\u7ecd\u4e86\u4f7f\u7528TMLE\uff08Targeted Maximum Likelihood\uff09\u65b9\u6cd5\u5728\u91cd\u75c7\u76d1\u62a4\u5ba4\u7814\u7a76\u4e2d\u7684\u5e94\u7528\u3002\u62a5\u544a\u63d0\u5230\u4e86\u76f8\u5173\u7684\u6570\u636e\u91cd\u7528\u95ee\u9898\uff0c\u8ba8\u8bba\u4e86\u56e0\u679c\u63a8\u65ad\u548c\u89c2\u5bdf\u6570\u636e\u5728\u836f\u7269\u5f00\u53d1\u548c\u4e34\u5e8a\u6307\u5357\u4e2d\u7684\u5e94\u7528\uff0c\u4ee5\u53ca\u5982\u4f55\u4f7f\u7528TMLE\u548c\u8d85\u7ea7\u5b66\u4e60\u8005\u6765\u63d0\u9ad8\u6548\u679c\u3002<\/span><\/p>\n

            \u521b\u65b0<\/span><\/strong><\/h3>\n
              \n
            1. \n

              \u4f7f\u7528TMLE\u65b9\u6cd5\u8fdb\u884c\u56e0\u679c\u63a8\u65ad\uff0c\u5f3a\u8c03TMLE\u65b9\u6cd5\u5728\u56e0\u679c\u63a8\u65ad\u4e2d\u7684\u5e94\u7528\uff0c\u4ee5\u63d0\u9ad8\u6570\u636e\u5206\u6790\u7684\u7cbe\u786e\u6027\u548c\u7a33\u5065\u6027\u3002<\/span><\/p>\n<\/li>\n

            2. \n

              \u8d85\u7ea7\u5b66\u4e60\u8005\u7684\u4f7f\u7528\uff0c\u4ecb\u7ecd\u4e86\u8d85\u7ea7\u5b66\u4e60\u8005\u7684\u6982\u5ff5\uff0c\u5373\u4f7f\u7528\u591a\u4e2a\u5b66\u4e60\u8005\u548c\u4ea4\u53c9\u9a8c\u8bc1\u6765\u63d0\u9ad8\u9884\u6d4b\u6548\u679c\uff0c\u7279\u522b\u662f\u5728\u5904\u7406\u590d\u6742\u6570\u636e\u65f6\u3002<\/span><\/p>\n<\/li>\n

            3. \n
              \u5bf9\u591a\u9879\u7814\u7a76\u7684\u5206\u6790\uff0c\u4f5c\u8005\u5bf9\u591a\u9879\u91cd\u75c7\u76d1\u62a4\u5ba4\u7814\u7a76\u7684\u7ed3\u679c\u8fdb\u884c\u4e86\u5206\u6790\u548c\u6bd4\u8f83\uff0c\u5f3a\u8c03\u4e86TML\u5728\u4e0d\u540c\u7814\u7a76\u4e2d\u7684\u5e94\u7528\u548c\u6548\u679c\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

              <\/span><\/section>\n
              \u6574\u7406\u8005\uff1a\u738b\u6587\u793c<\/span><\/section>\n

              <\/section>\n


              <\/span><\/p>\n

              \n
              \n
              \n
              \n
              \n

              <\/section>\n

              <\/section>\n
              \n
              <\/svg><\/section>\n<\/section>\n<\/section>\n
              \n

              \u4e0d\u662f\u968f\u673a\u7f3a\u5931\u7684<\/strong><\/strong><\/span><\/strong><\/strong><\/p>\n

              \u5e26\u6709\u4e2d\u4ecb\u53d8\u91cf\u548c\u7ed3\u679c\u7684\u4e2d\u4ecb\u5206\u6790<\/strong><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

              \n

              <\/section>\n

              <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


              <\/p>\n

              \n
              \u4e3b\u9898\uff1aMediation analysis with the mediator and outcome missing not at random<\/span><\/section>\n
              \u4e3b\u8bb2\u4eba\uff1aFan Yang<\/span><\/section>\n
              2022\u5e74\u81f3\u4eca\u6e05\u534e\u5927\u5b66\u6559\u6388\uff0c\u4e18\u6210\u6850\u6570\u5b66\u79d1\u5b66\u4e2d\u5fc3\u30022021-2022\u79d1\u7f57\u62c9\u591a\u5927\u5b66\u5b89\u8212\u8328\u533b\u5b66\u9662\u751f\u7269\u7edf\u8ba1\u5b66\u7cfb\u526f\u6559\u6388\u30022017-2021\u79d1\u7f57\u62c9\u591a\u5927\u5b66\u5b89\u8212\u8328\u533b\u5b66\u9662\u7edf\u8ba1\u5b66\u7cfb\u52a9\u7406\u6559\u6388\u30022014-2017\u516c\u5171\u536b\u751f\u79d1\u5b66\u7cfb\u52a9\u7406\u6559\u6388\uff0c\u829d\u52a0\u54e5\u5927\u5b66\u3002\u7814\u7a76\u9886\u57df\uff1a\u56e0\u679c\u5206\u6790\u3001\u6570\u636e\u7f3a\u5931\u3002<\/span><\/section>\n
              \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
              https:\/\/ymsc.tsinghua.edu.cn\/info\/1031\/2927.htm<\/span><\/section>\n<\/blockquote>\n


              <\/h2>\n

              \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
              Zuo, S., Ghosh, D., Ding, P., & Yang, F. (2022). Mediation analysis with the mediator and outcome missing not at random. arXiv preprint arXiv:2212.05577<\/em>.<\/span><\/section>\n


              <\/h2>\n

              \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

              \u6458\u8981<\/span><\/strong><\/h3>\n
              \u8c03\u89e3\u5206\u6790\u88ab\u5e7f\u6cdb\u7528\u4e8e\u7814\u7a76\u5bfc\u81f4\u67d0\u79cd\u6548\u5e94\u4ea7\u751f\u7684\u76f4\u63a5\u548c\u95f4\u63a5\u56e0\u679c\u8def\u5f84\u3002\u7136\u800c\uff0c\u8bb8\u591a\u8c03\u89e3\u5206\u6790\u7814\u7a76\u9762\u4e34\u4e2d\u4ecb\u53d8\u91cf\u548c\u7ed3\u679c\u53d8\u91cf\u7f3a\u5931\u7684\u6311\u6218\u3002\u901a\u5e38\u60c5\u51b5\u4e0b\uff0c\u5f53\u4e2d\u4ecb\u53d8\u91cf\u548c\u7ed3\u679c\u53d8\u91cf\u7684\u7f3a\u5931\u4e0d\u662f\u968f\u673a\u53d1\u751f\u65f6\uff0c\u8981\u60f3\u8bc6\u522b\u76f4\u63a5\u548c\u95f4\u63a5\u6548\u5e94\u5c31\u9700\u8981\u8fdb\u4e00\u6b65\u7684\u5047\u8bbe\u3002\u5728\u8fd9\u9879\u5de5\u4f5c\u4e2d\uff0c\u6211\u4eec\u7814\u7a76\u4e86\u5728\u4e00\u4e9b\u53ef\u89e3\u91ca\u7684\u673a\u5236\u4e0b\uff0c\u5141\u8bb8\u4e2d\u4ecb\u53d8\u91cf\u548c\u7ed3\u679c\u53d8\u91cf\u7684\u7f3a\u5931\u4e0d\u662f\u968f\u673a\u7684\u60c5\u51b5\u4e0b\uff0c\u76f4\u63a5\u548c\u95f4\u63a5\u6548\u5e94\u7684\u53ef\u8bc6\u522b\u6027\u3002\u6211\u4eec\u901a\u8fc7\u6a21\u62df\u7814\u7a76\u8bc4\u4f30\u4e86\u8fd9\u4e9b\u673a\u5236\u4e0b\u7684\u7edf\u8ba1\u63a8\u65ad\u6027\u80fd\uff0c\u5e76\u901a\u8fc7\u56fd\u5bb6\u5de5\u4f5c\u573a\u6240\u7814\u7a76\u6765\u8bf4\u660e\u6240\u63d0\u51fa\u7684\u65b9\u6cd5\u3002<\/span><\/section>\n

              <\/span><\/section>\n

              \u521b\u65b0<\/span><\/strong><\/h3>\n
              \u8fd9\u7bc7\u6587\u7ae0\u7684\u521b\u65b0\u70b9\u5728\u4e8e\u7814\u7a76\u4e86\u4e2d\u4ecb\u53d8\u91cf\u548c\u7ed3\u679c\u53d8\u91cf\u7f3a\u5931\u975e\u968f\u673a\uff08MNAR\uff09\u60c5\u51b5\u4e0b\u7684\u56e0\u679c\u4e2d\u4ecb\u6548\u5e94\u7684\u53ef\u8bc6\u522b\u6027\uff0c\u5e76\u63d0\u4f9b\u4e86\u5728\u5404\u79cd\u53ef\u89e3\u91ca\u7684MNAR\u5047\u8bbe\u4e0b\u7684\u8bc6\u522b\u6761\u4ef6\u3002\u8fd9\u4e00\u7814\u7a76\u586b\u8865\u4e86\u56e0\u679c\u63a8\u65ad\u9886\u57df\u4e2d\u7684\u4e00\u4e2a\u91cd\u8981\u7a7a\u767d\uff0c\u56e0\u4e3aMNAR\u60c5\u51b5\u901a\u5e38\u6bd4MAR\u60c5\u51b5\u66f4\u4e3a\u590d\u6742\uff0c\u9700\u8981\u66f4\u591a\u7684\u5047\u8bbe\u548c\u65b9\u6cd5\u6765\u8fdb\u884c\u5206\u6790\u3002\u867d\u7136\u4e4b\u524d\u7684\u7814\u7a76\u5df2\u7ecf\u5173\u6ce8\u4e86\u5728\u4e0d\u540c\u60c5\u5883\u4e0b\u7684\u4e00\u4e9b\u53ef\u8bc6\u522b\u6027\u95ee\u9898\uff0c\u4f46\u662f\u5bf9\u4e8e\u4e2d\u4ecb\u548c\u7ed3\u679c\u540c\u65f6\u4e3aMNAR\u7684\u60c5\u51b5\uff0c\u4ee5\u53ca\u4e2d\u4ecb\u53d8\u91cf\u7684\u7f3a\u5931\u95ee\u9898\uff0c\u4e4b\u524d\u7684\u7814\u7a76\u8f83\u5c11\uff0c\u8fd9\u7bc7\u6587\u7ae0\u901a\u8fc7\u63d0\u51fa\u65b0\u7684\u65b9\u6cd5\u548c\u7406\u8bba\u7ed3\u679c\u6765\u586b\u8865\u4e86\u8fd9\u4e00\u7814\u7a76\u7a7a\u767d\u3002\u6587\u7ae0\u8fd8\u901a\u8fc7\u6a21\u62df\u7814\u7a76\u548c\u5e94\u7528\u5230\u5b9e\u9645\u6570\u636e\u4e2d\u7684\u6848\u4f8b\u9a8c\u8bc1\u4e86\u63d0\u51fa\u65b9\u6cd5\u7684\u6027\u80fd\uff0c\u4e3a\u56e0\u679c\u4e2d\u4ecb\u6548\u5e94\u7684\u7814\u7a76\u63d0\u4f9b\u4e86\u6709\u529b\u7684\u5de5\u5177\u548c\u65b9\u6cd5\u3002<\/span><\/section>\n

              <\/section>\n
              \u6574\u7406\u8005\uff1a\u82cf\u5a49\u5a77<\/span><\/section>\n

              <\/section>\n


              <\/span><\/p>\n

              \n
              \n
              \n
              \n
              \n

              <\/section>\n

              <\/section>\n
              \n
              <\/svg><\/section>\n<\/section>\n<\/section>\n
              \n

              \u4e8c\u5143\u7ed3\u679c\u591a\u6548\u7a33\u5065\u5bf9\u6570\u7ebf\u6027\u6a21\u578b<\/strong><\/p>\n

              \u5b5f\u5fb7\u5c14\u968f\u673a\u5316\u5206\u6790<\/strong><\/p>\n<\/section>\n

              \n

              <\/section>\n

              <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


              <\/p>\n

              \n
              \u4e3b\u9898\uff1aMendelian randomization analysis with pleiotropy-robust log-linear models for binary outcomes<\/span><\/section>\n
              \u4e3b\u8bb2\u4eba\uff1aJinzhu Jia<\/span><\/section>\n
              \u5317\u4eac\u5927\u5b66\u516c\u5171\u536b\u751f\u5b66\u9662\u7814\u7a76\u5458\uff0c\u535a\u58eb\u751f\u5bfc\u5e08\u30022009\u5e741\u6708\u5317\u4eac\u5927\u5b66\u535a\u58eb\u6bd5\u4e1a\u30022009\u5e741\u6708\u81f32010\u5e7412\u6708\uff0cUC Berkeley \u535a\u58eb\u540e\u30022011\u5e741\u6708\u81f32018\u5e741\u6708\u4efb\u804c\u4e8e\u5317\u4eac\u5927\u5b66\u6570\u5b66\u79d1\u5b66\u5b66\u9662\u6982\u7387\u7edf\u8ba1\u7cfb\u548c\u5317\u4eac\u5927\u5b66\u7edf\u8ba1\u4e2d\u5fc3\uff0c\u671f\u95f4\u8bbf\u95ee\u54c8\u4f5b\u5927\u5b66\u4e00\u5e74\u30022018 \u5e742\u6708\u52a0\u5165\u5317\u4eac\u5927\u5b66\u516c\u5171\u536b\u751f\u5b66\u9662\u3002\u4e3b\u8981\u7814\u7a76\u65b9\u5411\u662f\u9ad8\u7ef4\u7edf\u8ba1\u63a8\u65ad\u3001\u5927\u6570\u636e\u5206\u6790\u3001\u7edf\u8ba1\u673a\u5668\u5b66\u4e60\u3001\u56e0\u679c\u63a8\u65ad\u3001\u751f\u7269\u7edf\u8ba1\u7b49\u3002\u5728\u53d8\u91cf\u9009\u62e9\u65b9\u6cd5\u7684\u7406\u8bba\u7814\u7a76\u3001\u9ad8\u7ef4\u6570\u636e\u548c\u5927\u6570\u636e\u7edf\u8ba1\u5b66\u4e60\u7684\u5e94\u7528\u4ee5\u53ca\u56e0\u679c\u63a8\u65ad\u7b49\u9886\u57df\u53d1\u8868\u8bba\u6587\u591a\u7bc7\u3002\u62c5\u4efb\u4e2d\u56fd\u6982\u7387\u7edf\u8ba1\u7edf\u8ba1\u5b66\u4f1a\u526f\u79d8\u4e66\u957f\u3001\u9752\u5e74\u7edf\u8ba1\u5b66\u5bb6\u534f\u4f1a\u5e38\u52a1\u7406\u4e8b\u3001\u73b0\u573a\u7edf\u8ba1\u7814\u7a76\u4f1a\u8ba1\u7b97\u7edf\u8ba1\u5206\u4f1a\u7406\u4e8b\u3001\u73b0\u573a\u7edf\u8ba1\u7814\u7a76\u4f1a\u9ad8\u7ef4\u6570\u636e\u7edf\u8ba1\u4f1a\u7406\u4e8b\u3002<\/span><\/section>\n
              \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/sph.pku.edu.cn\/info\/1566\/4001.htm<\/span><\/section>\n<\/blockquote>\n


              <\/span><\/h2>\n

              \u5f15\u7528\u6587\u732e\uff1a<\/span><\/strong><\/span><\/h2>\n
              Li M, Huang T, Jia J. Mendelian randomization analysis with pleiotropy-robust log-linear model for binary outcomes[J]. bioRxiv, 2023: 2023.06. 03.543587.<\/span><\/section>\n


              <\/h2>\n

              \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

              \u6458\u8981<\/span><\/h3>\n
              \u5b5f\u5fb7\u5c14\u968f\u673a\u5316(Mendelian randomization, MR)\u662f\u4e00\u79cd\u5229\u7528\u9057\u4f20\u53d8\u5f02\u4f5c\u4e3a\u5de5\u5177\u53d8\u91cf\u6765\u63a8\u65ad\u6027\u72b6\u4e4b\u95f4\u56e0\u679c\u5173\u7cfb\u7684\u7edf\u8ba1\u6280\u672f\u3002\u5728\u5904\u7406\u4e8c\u5143\u7ed3\u679c\u65f6\uff0c\u6709\u6548\u7684MR\u5206\u6790\u6709\u4e24\u4e2a\u5177\u6709\u6311\u6218\u6027\u7684\u969c\u788d\uff0c\u5373\u4f20\u7edf\u6bd4\u7387\u4f30\u8ba1\u5668\u7684\u4e0d\u4e00\u81f4\u6027\u548c\u6c34\u5e73\u591a\u6548\u6027\u7684\u5b58\u5728\u3002\u6700\u8fd1\u7684MR\u65b9\u6cd5\u4e3b\u8981\u96c6\u4e2d\u5728\u901a\u8fc7\u6c47\u603b\u7edf\u8ba1\u6765\u5904\u7406\u591a\u6548\u6027\u3002\u6211\u4eec\u63d0\u51fa\u4e86\u4e24\u79cd\u65b0\u9896\u7684\u57fa\u4e8e\u4e2a\u4f53\u6570\u636e\u7684\u65b9\u6cd5\uff0c\u5206\u522b\u547d\u540d\u4e3a\u968f\u673a\u6548\u5e94(random-effects)\u548c\u56fa\u5b9a\u6548\u5e94(fixed-effects) mr – prolim\uff0c\u4ee5\u514b\u670d\u8fd9\u4e24\u4e2a\u969c\u788d\u3002\u8fd9\u4e24\u79cd\u65b9\u6cd5\u91c7\u7528\u98ce\u9669\u6bd4(RR)\u6765\u5b9a\u4e49\u8fde\u7eed\u6216\u4e8c\u5143\u66b4\u9732\u7684\u56e0\u679c\u6548\u5e94\u3002\u968f\u673a\u6548\u5e94MR-PROLLIM\u6a21\u578b\u4e0e\u591a\u6548\u6027\u76f8\u5173\uff0c\u89e3\u91ca\u4e86\u53d8\u91cf\u9009\u62e9\uff0c\u5e76\u5141\u8bb8\u4f7f\u7528\u8f83\u5f31\u7684\u5de5\u5177\u3002\u56fa\u5b9a\u6548\u5e94MR-PROLLIM\u53ea\u80fd\u4f7f\u7528\u5c11\u6570\u51e0\u4e2a\u9009\u5b9a\u7684\u53d8\u4f53\u3002\u672c\u6587\u8bc1\u660e\uff0c\u968f\u673a\u6548\u5e94MR-PROLLIM\u8868\u73b0\u51fa\u5f88\u9ad8\u7684\u7edf\u8ba1\u80fd\u529b\uff0c\u540c\u65f6\u6bd4\u5176\u7ade\u4e89\u5bf9\u624b\u4ea7\u751f\u66f4\u5c11\u7684\u5047\u9633\u6027\u68c0\u6d4b\u3002\u56fa\u5b9a\u6548\u5e94MR-PROLLIM\u7684\u8868\u73b0\u901a\u5e38\u4ecb\u4e8e\u7ecf\u5178\u7684\u4e2d\u4f4d\u6570\u548c\u4f17\u6570\u4f30\u8ba1\u5668\u4e4b\u95f4\u3002\u5728\u6211\u4eec\u7684\u82f1\u56fd\u751f\u7269\u94f6\u884c\u6570\u636e\u5206\u6790\u4e2d\uff0c\u6211\u4eec\u8fd8\u53d1\u73b0(i) MR\u6bd4\u7387\u65b9\u6cd5\u5f80\u5f80\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u4f4e\u4f30\u4e86\u4e8c\u5143\u66b4\u9732\u6548\u5e94;(2)\u7ea626.5%\u7684\u6027\u72b6\u5bf9\u5b58\u5728\u663e\u8457\u7684\u591a\u6548\u76f8\u5173;(iii)\u591a\u6548\u654f\u611f\u6cd5\u5bf9\u975e\u96f6\u6548\u5e94\u7684\u4f30\u8ba1\u76f8\u5bf9\u504f\u5dee\u4e3a-103.7% ~ 178.0%\u3002MR- prollim\u663e\u793a\u4e86\u4fc3\u8fdb\u5bf9\u4e8c\u5143\u7ed3\u679c\u8fdb\u884c\u66f4\u4e25\u683c\u3001\u66f4\u7a33\u5065\u7684MR\u5206\u6790\u7684\u6f5c\u529b\u3002<\/span><\/section>\n


              <\/h3>\n

              \u521b\u65b0<\/span><\/strong><\/span><\/h3>\n
                \n
              1. \n

                \u5904\u7406\u4e8c\u5143\u7ed3\u679c\u7684\u65b0\u65b9\u6cd5\uff1a\u8be5\u62a5\u544a\u4ecb\u7ecd\u4e86\u4e00\u79cd\u65b0\u7684\u65b9\u6cd5\uff0c\u7528\u4e8e\u5904\u7406\u4e8c\u5143\u7ed3\u679c\u7684\u56e0\u679c\u5173\u7cfb\u4f30\u8ba1\u3002\u4f20\u7edf\u7684\u65b9\u6cd5\u901a\u5e38\u4f7f\u7528\u903b\u8f91\u56de\u5f52\u6a21\u578b\uff0c\u800c\u62a5\u544a\u4e2d\u63d0\u51fa\u4e86\u4f7f\u7528\u5bf9\u6570\u7ebf\u6027\u6a21\u578b\u7684\u65b9\u6cd5\uff0c\u4ee5\u83b7\u5f97\u66f4\u4e00\u81f4\u7684\u4f30\u8ba1\u7ed3\u679c\u3002<\/span><\/p>\n<\/li>\n

              2. \n

                \u9ad8\u7ef4\u57fa\u56e0\u6570\u636e\u7684\u5904\u7406\uff1a\u7814\u7a76\u4e2d\u6d89\u53ca\u4e86\u591a\u4e2a\u57fa\u56e0\u4f4d\u70b9\uff08SNPs\uff09\uff0c\u5e76\u4ecb\u7ecd\u4e86\u5982\u4f55\u5c06\u5b83\u4eec\u5408\u5e76\u6210\u9ad8\u7ef4\u4eea\u5668\u53d8\u91cf\u4ee5\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u3002\u8fd9\u79cd\u65b9\u6cd5\u53ef\u4ee5\u63d0\u9ad8\u7edf\u8ba1\u529f\u6548\u5e76\u66f4\u597d\u5730\u6355\u6349\u57fa\u56e0\u53d8\u5f02\u4e0e\u611f\u5174\u8da3\u7ed3\u679c\u4e4b\u95f4\u7684\u590d\u6742\u5173\u7cfb\u3002<\/span><\/p>\n<\/li>\n

              3. \n

                \u5904\u7406\u4eea\u5668\u53d8\u91cf\u7684\u4e0d\u4e00\u81f4\u6027\uff1a\u4f20\u7edf\u7684\u4e24\u9636\u6bb5\u4f30\u8ba1\u5668\u5728\u5904\u7406\u4e8c\u5143\u7ed3\u679c\u65f6\u53ef\u80fd\u4e0d\u4e00\u81f4\uff0c\u8be5\u62a5\u544a\u63d0\u4f9b\u4e86\u4e00\u79cd\u65b9\u6cd5\u6765\u5904\u7406\u8fd9\u79cd\u4e0d\u4e00\u81f4\u6027\uff0c\u4ece\u800c\u66f4\u51c6\u786e\u5730\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u3002<\/span><\/p>\n<\/li>\n

              4. \n

                \u5904\u7406\u5de5\u5177\u53d8\u91cf\u7684\u6c34\u5e73\u6df7\u6742\u6027\uff1a\u62a5\u544a\u4ecb\u7ecd\u4e86\u5982\u4f55\u533a\u5206\u6709\u6548\u7684\u5de5\u5177\u53d8\u91cf\u548c\u5b58\u5728\u6c34\u5e73\u6df7\u6742\u6027\u7684\u5de5\u5177\u53d8\u91cf\uff0c\u4ee5\u63d0\u9ad8\u56e0\u679c\u4f30\u8ba1\u7684\u53ef\u9760\u6027\u3002<\/span><\/p>\n<\/li>\n

              5. \n
                \u5728\u771f\u5b9e\u6570\u636e\u5206\u6790\u4e2d\u7684\u9a8c\u8bc1\uff1a\u7814\u7a76\u4e2d\u8fdb\u884c\u4e86\u5927\u91cf\u7684\u771f\u5b9e\u6570\u636e\u5206\u6790\uff0c\u4ee5\u9a8c\u8bc1\u65b0\u65b9\u6cd5\u7684\u6027\u80fd\uff0c\u7ed3\u679c\u8868\u660e\u65b0\u65b9\u6cd5\u80fd\u591f\u6709\u6548\u5730\u68c0\u6d4b\u51fa\u56e0\u679c\u5173\u7cfb\uff0c\u76f8\u5bf9\u4e8e\u4f20\u7edf\u65b9\u6cd5\u5177\u6709\u66f4\u9ad8\u7684\u7edf\u8ba1\u529f\u6548\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

                <\/section>\n
                \u6574\u7406\u8005\uff1a\u738b\u6587\u793c<\/span><\/section>\n


                <\/span><\/h1>\n

                <\/span><\/section>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u5206\u5272\u533a\u8bbe\u8ba1\u4e2d\u7684\u968f\u673a\u5316\u4e0e\u534f\u53d8\u91cf\u8c03\u6574<\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aRandomization and Covariate Adjustment in Split plot Designs<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aHanzhong Liu<\/span><\/section>\n
                \u6e05\u534e\u5927\u5b66\u7edf\u8ba1\u79d1\u5b66\u4e2d\u5fc3\u526f\u6559\u6388\uff082018\/12\u81f3\u4eca\uff09\uff1b\u6b64\u524d\uff0c\u4ed6\u5728\u540c\u4e00\u4e2d\u5fc3\u62c5\u4efb\u52a9\u7406\u6559\u6388\uff082016\/08 – 2018\/12\uff09\uff0c\u5e76\u66fe\u5728\u52a0\u5229\u798f\u5c3c\u4e9a\u5927\u5b66\u4f2f\u514b\u5229\u5206\u6821\u4f5c\u4e3a\u535a\u58eb\u540e\u5b66\u8005\uff082014\/07 – 2016\/06\uff09\u3002\u4ed6\u5728\u5317\u4eac\u5927\u5b66\uff082009\/09 – 2014\/06\uff09\u548c\u4e2d\u56fd\u79d1\u5b66\u6280\u672f\u5927\u5b66\uff082005\/09 – 2009\/06\uff09\u83b7\u5f97\u5b66\u4f4d\u3002\u57282012\/09\u81f32014\/04\u671f\u95f4\uff0c\u4ed6\u66fe\u524d\u5f80\u52a0\u5229\u798f\u5c3c\u4e9a\u5927\u5b66\u4f2f\u514b\u5229\u5206\u6821\u4f5c\u4e3a\u8bbf\u95ee\u5b66\u8005\u3002\u7814\u7a76\u5174\u8da3\u5305\u62ec\u56e0\u679c\u63a8\u65ad\u3001\u9ad8\u7ef4\u7edf\u8ba1\u3001\u5927\u6570\u636e\u548c\u673a\u5668\u5b66\u4e60\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                https:\/\/tsing.v-dk.com\/teambuilder\/faculty\/hanzhong-liu\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Shi, Wenqi, Anqi Zhao, and Hanzhong Liu. “Rerandomization and covariate adjustment in split-plot designs.” arXiv preprint arXiv:2209.12385<\/em> (2022).<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u672c\u6587\u63a2\u8ba8\u4e86\u6e90\u4e8e\u519c\u4e1a\u79d1\u5b66\u7684split-plot\u8bbe\u8ba1\uff0c\u5176\u4e2d\u5b9e\u9a8c\u5355\u4f4d\uff08\u4ea6\u79f0\u4e3a\u5b50\u56fe\uff09\u5d4c\u5957\u5728\u79f0\u4e3a\u6574\u4f53\u56fe\u7684\u7ec4\u5185\u3002\u8be5\u8bbe\u8ba1\u5728\u6574\u4f53\u56fe\u5c42\u9762\u901a\u8fc7\u96c6\u7fa4\u968f\u673a\u5316\u5206\u914d\u6574\u4f53\u56fe\u5e72\u9884\uff0c\u5e76\u5728\u5b50\u56fe\u5c42\u9762\u901a\u8fc7\u5206\u5c42\u968f\u673a\u5316\u5206\u914d\u5b50\u56fe\u5e72\u9884\u3002\u968f\u673a\u5316\u673a\u5236\u4fdd\u8bc1\u4e86\u5728\u6574\u4f53\u56fe\u548c\u5b50\u56fe\u5c42\u9762\u7684\u5e73\u5747\u534f\u53d8\u91cf\u5e73\u8861\uff0c\u5e76\u901a\u8fc7Horvitz-Thompson\u548cHajek\u4f30\u8ba1\u5668\u786e\u4fdd\u4e86\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u7684\u4e00\u81f4\u63a8\u65ad\u3002\u7136\u800c\uff0c\u5728\u6709\u9650\u7684\u6837\u672c\u4e2d\u7ecf\u5e38\u51fa\u73b0\u534f\u53d8\u91cf\u5931\u8861\uff0c\u53ef\u80fd\u5bfc\u81f4\u63a8\u65ad\u7684\u5927\u53d8\u5f02\u6027\u548c\u6761\u4ef6\u504f\u5dee\u3002\u91cd\u65b0\u968f\u673a\u5316\u5728\u968f\u673a\u5b9e\u9a8c\u7684\u8bbe\u8ba1\u9636\u6bb5\u88ab\u5e7f\u6cdb\u4f7f\u7528\u4ee5\u6539\u5584\u534f\u53d8\u91cf\u5e73\u8861\u3002\u73b0\u6709\u7684\u91cd\u65b0\u968f\u673a\u5316\u6587\u732e\u4e3b\u8981\u96c6\u4e2d\u5728\u6cbb\u7597\u53ea\u5728\u5355\u5143\u6216\u7ec4\u5c42\u9762\u5206\u914d\u7684\u8bbe\u8ba1\u4e0a\uff0c\u800c\u4e0d\u662f\u4e24\u8005\u90fd\u6709\uff0c\u8fd9\u4f7f\u5f97split-plot\u8bbe\u8ba1\u4e2d\u7684\u91cd\u65b0\u968f\u673a\u5316\u7406\u8bba\u4ecd\u662f\u4e00\u4e2a\u672a\u89e3\u51b3\u7684\u95ee\u9898\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u672c\u6587\u9996\u6b21\u63d0\u51fa\u4e86\u57fa\u4e8eMahalanobis\u8ddd\u79bb\u8fdb\u884csplit-plot\u8bbe\u8ba1\u4e2d\u91cd\u65b0\u968f\u673a\u5316\u7684\u4e24\u79cd\u7b56\u7565\uff0c\u5e76\u5efa\u7acb\u4e86\u76f8\u5e94\u7684\u57fa\u4e8e\u8bbe\u8ba1\u7684\u7406\u8bba\u3002\u7814\u7a76\u8868\u660e\uff0c\u91cd\u65b0\u968f\u673a\u5316\u53ef\u4ee5\u63d0\u9ad8Horvitz-Thompson\u548cHajek\u4f30\u8ba1\u5668\u7684\u6e10\u8fd1\u6548\u7387\u3002\u6b64\u5916\uff0c\u8fd8\u63d0\u51fa\u4e86\u4e24\u79cd\u5206\u6790\u9636\u6bb5\u7684\u534f\u53d8\u91cf\u8c03\u6574\u65b9\u6cd5\uff0c\u4e0e\u91cd\u65b0\u968f\u673a\u5316\u7ed3\u5408\u65f6\uff0c\u8fd9\u4e9b\u65b9\u6cd5\u53ef\u4ee5\u8fdb\u4e00\u6b65\u63d0\u9ad8\u6e10\u8fd1\u6548\u7387\u3002\u901a\u8fc7\u6570\u503c\u7814\u7a76\uff0c\u8bc1\u660e\u4e86\u6240\u63d0\u65b9\u6cd5\u7684\u6709\u6548\u6027\u548c\u6539\u8fdb\u6548\u7387\u3002<\/span><\/section>\n

                <\/section>\n
                \u6574\u7406\u8005\uff1a\u51af\u5e06<\/span><\/section>\n


                <\/h1>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u81ea\u52a8\u5de5\u5177\u53d8\u91cf\u9009\u62e9\u548c\u6c34\u5e73\u591a\u6548\u5e94\u5efa\u6a21\u4e0b<\/strong>\u7684<\/strong><\/p>\n

                \u57fa\u4e8e\u4f3c\u7136\u7684\u5b5f\u5fb7\u5c14\u968f\u673a\u5316\u5206\u6790<\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aLikelihood based Mendelian randomization analysis with automated instrument selection and horizontal pleiotropic modeling<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aZhongshan Yuan\uff0c\u5c71\u4e1c\u5927\u5b66\u751f\u7269\u7edf\u8ba1\u5b66\u9662\u6559\u6388\uff0c\u535a\u5bfc\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                http:\/\/www.en.sph.sdu.edu.cn\/info\/1484\/1155.htm<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

                Yuan, Z., Liu, L., Guo, P., Yan, R., Xue, F., & Zhou, X. (2021). Likelihood based Mendelian randomization analysis with automated instrument selection and horizontal pleiotropic modeling. https:\/\/doi.org\/10.1101\/2021.11.03.21265848<\/span><\/p>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u5b5f\u5fb7\u5c14\u968f\u673a\u5316(MR)\u662f\u8bc6\u522b\u6f5c\u5728\u75be\u75c5\u7684\u56e0\u679c\u5371\u9669\u56e0\u7d20\u7684\u5e38\u7528\u5de5\u5177\u3002\u5728\u8fd9\u91cc\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u6709\u6548\u7684MR\u5206\u6790\u65b9\u6cd5\uff0cMRAID\u3002MRAID\u501f\u9274\u7cbe\u7ec6\u6620\u5c04\u5206\u6790\u7684\u601d\u60f3\uff0c\u5bf9\u5f7c\u6b64\u4e4b\u95f4\u6f5c\u5728\u7684\u9ad8\u5ea6\u8fde\u9501\u4e0d\u5e73\u8861\u7684\u5019\u9009snp\u521d\u59cb\u96c6\u8fdb\u884c\u5efa\u6a21\uff0c\u5e76\u81ea\u52a8\u4ece\u4e2d\u9009\u62e9\u5408\u9002\u7684\u5de5\u5177\u8fdb\u884c\u56e0\u679c\u63a8\u7406\u3002MRAID\u8fd8\u660e\u786e\u5730\u6a21\u62df\u4e86\u5728\u590d\u6742\u6027\u72b6\u5206\u6790\u4e2d\u5e7f\u6cdb\u5b58\u5728\u7684\u4e0d\u76f8\u5173\u548c\u76f8\u5173\u6c34\u5e73\u591a\u6548\u6548\u5e94\u3002MRAID\u901a\u8fc7\u8054\u5408\u4f3c\u7136\u6846\u67b6\u5b9e\u73b0\u8fd9\u4e24\u9879\u4efb\u52a1\uff0c\u5e76\u4f9d\u8d56\u4e8e\u53ef\u6269\u5c55\u7684\u57fa\u4e8e\u91c7\u6837\u7684\u7b97\u6cd5\u6765\u8ba1\u7b97\u6821\u51c6\u7684p\u503c\u3002\u5168\u9762\u548c\u73b0\u5b9e\u7684\u4eff\u771f\u8868\u660e\uff0cMRAID\u53ef\u4ee5\u63d0\u4f9b\u6821\u51c6\u7684I\u578b\u8bef\u5dee\u63a7\u5236\uff0c\u51cf\u5c11\u8bef\u62a5\uff0c\u540c\u65f6\u6bd4\u73b0\u6709\u65b9\u6cd5\u66f4\u5f3a\u5927\u3002\u6211\u4eec\u8bf4\u660e\u4e86MRAID\u5728\u82f1\u56fd\u751f\u7269\u94f6\u884c645\u5bf9\u7279\u5f81\u5bf9\u7684MR\u7b5b\u9009\u5206\u6790\u4e2d\u7684\u597d\u5904\uff0c\u786e\u5b9a\u4e86\u5fc3\u8840\u7ba1\u75be\u75c5\u76f8\u5173\u7279\u5f81\u7684\u591a\u79cd\u751f\u6d3b\u65b9\u5f0f\u56e0\u679c\u98ce\u9669\u56e0\u7d20\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u63d0\u51fa\u4e86\u4e00\u79cd\u57fa\u4e8e\u4f3c\u7136\u7684\u53cc\u6837\u672c MR \u65b9\u6cd5\u8fdb\u884c\u56e0\u679c\u63a8\u7406\u3002\u5177\u4f53\u6765\u8bf4\uff0c\u4f5c\u8005\u7684\u65b9\u6cd5\u6a21\u62df\u4e86\u4e00\u7ec4\u521d\u59cb\u7684\u5019\u9009SNP\u4eea\u5668\uff0c\u8fd9\u4e9b\u4eea\u5668\u5f7c\u6b64\u9ad8LD\uff0c\u5e76\u5728\u5b83\u4eec\u4e2d\u81ea\u52a8\u9009\u62e93\u4e2a\u4f5c\u4e3aMR\u5206\u6790\u7684\u5408\u9002\u4eea\u5668\u3002\u6b64\u5916\uff0c\u5728\u4f3c\u7136\u6846\u67b6\u4e2d\u8003\u8651\u4e86\u4e24\u79cd\u7c7b\u578b\u7684\u6c34\u5e73\u8d8b\u5411\u6027\uff0c\u5e76\u4e14\u4f9d\u8d56\u4e8e\u57fa\u4e8e\u53ef\u6269\u5c55\u7684\u57fa\u4e8e\u91c7\u6837\u7684\u7b97\u6cd5\u8fdb\u884c\u6821\u51c6\u7684 p \u503c\u8ba1\u7b97\u3002\u4f5c\u8005\u5c06\u6b64\u65b9\u6cd5\u79f0\u4e3a\u5177\u6709\u81ea\u52a8\u4eea\u5668\u786e\u5b9a (MRAID) \u7684\u53cc\u6837\u672c\u5b5f\u5fb7\u5c14\u968f\u673a\u5316\u3002\u4f5c\u8005\u901a\u8fc7\u5168\u9762\u548c\u73b0\u5b9e\u7684\u6a21\u62df\u8bc1\u660e\u4e86 MRAID8 \u7684\u6709\u6548\u6027\u3002\u4f5c\u8005\u8fd8\u5c06 MRAID \u5e94\u7528\u4e8e\u82f1\u56fd\u751f\u7269\u94f6\u884c\u4e2d 645 \u4e2a\u7279\u5f81\u5bf9\u7684 MR9 \u7b5b\u67e5\u5206\u6790\uff0c\u8bc6\u522b\u53ef\u80fd\u5f71\u54cd\u5fc3\u8840\u7ba1\u75be\u75c5\u76f8\u5173\u7279\u5f81\u7684\u751f\u6d3b\u65b9\u5f0f 10 \u5371\u9669\u56e0\u7d20\u3002<\/span><\/section>\n

                <\/span><\/section>\n
                \u6574\u7406\u8005\uff1a<\/span>\u6768\u660e\u54f2<\/span><\/span><\/section>\n


                <\/h1>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u56de\u5f52\u95f4\u65ad\u8bbe\u8ba1\u4e0b\u56e0\u679c\u63a8\u65ad\u7684\u4e9a\u7fa4\u4f53\u9009\u62e9<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aSelecting Subpopulations for Causal Inference in Regression Discontinuity Designs<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aFabrizia Mealli\uff0c\u610f\u5927\u5229\u4f5b\u7f57\u4f26\u8428\u5927\u5b66\u7edf\u8ba1\u4e0e\u8ba1\u7b97\u673a\u79d1\u5b66\u5b66\u9662\u6559\u6388<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                http:\/\/local.disia.unifi.it\/mealli\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Forastiere, L., Mealli, F., Pescarini, JuliaM., Fiaccone, RosemeireL., Barreto, MauricioL., & Mattei, A. (2022). Selecting Subpopulations for Causal Inference in Regression Discontinuity Designs<\/em>.<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u5df4\u897f\u5bb6\u5ead\u8865\u52a9\u8ba1\u5212(BF)\u662f\u4e00\u9879\u6709\u6761\u4ef6\u7684\u73b0\u91d1\u8f6c\u79fb\u8ba1\u5212\uff0c\u65e8\u5728\u901a\u8fc7\u76f4\u63a5\u73b0\u91d1\u8f6c\u79fb\u51cf\u5c11\u77ed\u671f\u8d2b\u56f0\uff0c\u5e76\u901a\u8fc7\u589e\u52a0\u5df4\u897f\u7a77\u4eba\u7684\u4eba\u529b\u8d44\u672c\u6765\u5bf9\u6297\u957f\u671f\u8d2b\u56f0\u3002\u83b7\u5f97Bolsa Familia\u798f\u5229\u7684\u8d44\u683c\u53d6\u51b3\u4e8e\u4e00\u4e2a\u622a\u6b62\u89c4\u5219\uff0c\u8be5\u89c4\u5219\u5c06BF\u7814\u7a76\u5206\u7c7b\u4e3a\u56de\u5f52\u95f4\u65ad(RD)\u8bbe\u8ba1\u3002\u4eceRD\u7814\u7a76\u4e2d\u63d0\u53d6\u56e0\u679c\u4fe1\u606f\u5177\u6709\u6311\u6218\u6027\u3002\u7ee7Li\u7b49\u4eba(2015)\u548cBranson\u548cMealli(2019)\u4e4b\u540e\uff0c\u6211\u4eec\u6b63\u5f0f\u5c06BF RD\u8bbe\u8ba1\u63cf\u8ff0\u4e3a\u6f5c\u5728\u7ed3\u679c\u65b9\u6cd5\u4e2d\u7684\u5c40\u90e8\u968f\u673a\u5b9e\u9a8c\u3002\u5728\u6b64\u6846\u67b6\u4e0b\uff0c\u53ef\u4ee5\u5728\u5c40\u90e8\u91cd\u53e0\u5047\u8bbe\u3001\u5c40\u90e8SUTVA\u548c\u5c40\u90e8\u53ef\u5ffd\u7565\u6027\u5047\u8bbe\u6210\u7acb\u7684\u4e9a\u4eba\u7fa4\u4e0a\u786e\u5b9a\u548c\u4f30\u8ba1\u56e0\u679c\u5f71\u54cd\u3002\u4e0e\u57fa\u4e8e\u8fde\u7eed\u6027\u5047\u8bbe\u7684\u5c40\u90e8\u56de\u5f52\u65b9\u6cd5\u76f8\u6bd4\uff0c\u8be5\u6846\u67b6\u7684\u6f5c\u5728\u4f18\u52bf\u5728\u4e8e\u56e0\u679c\u4f30\u8ba1\u3001\u5206\u6790\u548c\u7ed3\u679c\u89e3\u91ca\u3002\u8fd9\u79cd\u5c40\u90e8\u968f\u673a\u5316\u65b9\u6cd5\u7684\u4e00\u4e2a\u5173\u952e\u95ee\u9898\u662f\u5982\u4f55\u9009\u62e9\u6211\u4eec\u53ef\u4ee5\u5f97\u51fa\u6709\u6548\u56e0\u679c\u63a8\u65ad\u7684\u4e9a\u7fa4\u4f53\u3002\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u57fa\u4e8e\u8d1d\u53f6\u65af\u6a21\u578b\u7684\u6709\u9650\u6df7\u5408\u65b9\u6cd5\u6765\u805a\u7c7b\uff0c\u5c06\u89c2\u6d4b\u6570\u636e\u5206\u7c7b\u4e3ad\u5047\u8bbe\u6210\u7acb\u548c\u4e0d\u6210\u7acb\u7684\u4e9a\u79cd\u7fa4\u3002\u8fd9\u79cd\u65b9\u6cd5\u6709\u91cd\u8981\u7684\u4f18\u70b9:a)\u5b83\u53ef\u4ee5\u8003\u8651\u901a\u5e38\u88ab\u5ffd\u7565\u7684\u4e9a\u79cd\u7fa4\u6210\u5458\u7684\u4e0d\u786e\u5b9a\u6027;B)\u5b83\u5bf9\u4e9a\u79cd\u7fa4\u7684\u5f62\u72b6\u6ca1\u6709\u4efb\u4f55\u9650\u5236;C)\u53ef\u6269\u5c55\u5230\u9ad8\u7ef4\u8bbe\u7f6e;E)\u5b83\u5141\u8bb8\u9488\u5bf9\u66ff\u4ee3\u56e0\u679c\u4f30\u8ba1\u800c\u4e0d\u662f\u5e73\u5747\u6548\u5e94;f)\u5b83\u5bf9\u8fd0\u884c\u53d8\u91cf\u7684\u4e00\u5b9a\u7a0b\u5ea6\u7684\u64cd\u4f5c\/\u9009\u62e9\u5177\u6709\u9c81\u68d2\u6027\u3002\u6211\u4eec\u5e94\u7528\u6211\u4eec\u63d0\u51fa\u7684\u65b9\u6cd5\u6765\u8bc4\u4f30BF\u9879\u76ee\u5bf92009\u5e74\u9ebb\u98ce\u75c5\u53d1\u75c5\u7387\u7684\u56e0\u679c\u5f71\u54cd\uff0c\u5bf9\u8c61\u662f2007-2008\u5e74\u9996\u6b21\u5728\u5df4\u897f\u56fd\u5bb6\u793e\u4f1a\u9879\u76ee\u767b\u8bb0\u5904\u767b\u8bb0\u7684\u5df4\u897f\u5bb6\u5ead<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u4f5c\u8005\u5c06\u6b64\u5c40\u90e8\u968f\u673a\u5316\u6846\u67b6\u4e0e\u6211\u4eec\u65b0\u63d0\u51fa\u7684\u57fa\u4e8e\u8d1d\u53f6\u65af\u6a21\u578b\u7684\u6df7\u5408\u65b9\u6cd5\u76f8\u7ed3\u5408\uff0c\u4ee5\u8bc4\u4f30 Borsa Fam \u0301\u0131lia \u7a0b\u5e8f\u5bf9 leprosy \u53d1\u75c5\u7387\u7684\u56e0\u679c\u5f71\u54cd\u3002Borsa Fam \u0301\u0131lia \u7814\u7a76\u662f\u4e00\u9879\u9ad8\u7ef4\u7814\u7a76\uff0c\u5305\u62ec\u5927\u91cf\u5bb6\u5ead\u7684\u4fe1\u606f\uff0c\u8fd9\u610f\u5473\u7740\u8bc1\u4f2a\u6d4b\u8bd5\u4e0d\u80fd\u7528\u4e8e\u9009\u62e9\u5408\u9002\u7684\u4e9a\u7fa4\u8fdb\u884c\u6709\u6548\u7684\u56e0\u679c\u63a8\u7406\u3002\u6b64\u5916\uff0c\u5728\u6837\u672c\u4e2d\uff0c\u53ea\u6709 424 \u4e2a\u6848\u4f8b\u8d85\u8fc7 152 602 \u4e2a\u5bb6\u5ead\uff0c\u611f\u5174\u8da3\u7684\u7ed3\u679c leprosy \u53d1\u75c5\u7387\u5f88\u5c11\u89c1\u3002\u7ed3\u679c\u6b64\u529f\u80fd\u5728 RD \u8bbe\u8ba1\u4e2d\u7279\u522b\u5177\u6709\u6311\u6218\u6027\u3002\u4f7f\u7528\u4f5c\u8005\u7684\u57fa\u4e8e\u8d1d\u53f6\u65af\u6a21\u578b\u7684\u6df7\u5408\u65b9\u6cd5\u8fdb\u884c\u7814\u7a76\u7684\u8bbe\u8ba1\u548c\u5206\u6790\u9636\u6bb5\u4f7f\u6211\u4eec\u80fd\u591f\u9762\u5bf9\u8fd9\u4e24\u4e2a\u95ee\u9898\u3002\u7279\u522b\u662f\uff0c\u4f5c\u8005\u4e13\u6ce8\u4e8e\u6839\u636e\u7ed3\u679c\u3001\u5f3a\u5236\u53d8\u91cf\u548c\u534f\u53d8\u91cf\u7684\u540e\u9a8c\u5206\u5e03\u6765\u8bc6\u522b\u8fd9\u79cd\u56e0\u679c\u6548\u5e94\u7684\u5355\u4f4d\u5b50\u96c6\u7684\u5c40\u90e8\u6709\u9650\u6837\u672c\u56e0\u679c\u76f8\u5bf9\u98ce\u9669\u8fdb\u884c\u63a8\u65ad\u3002<\/span><\/section>\n

                <\/span><\/section>\n
                \u6574\u7406\u8005\uff1a<\/span>\u6768\u660e\u54f2<\/span><\/span><\/section>\n


                <\/h1>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u975e\u5bf9\u79f0\u6548\u7528\u51fd\u6570\u7684\u653f\u7b56\u5b66\u4e60<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aPolicy learning with asymmetric utilities<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aZhichao Jiang<\/span><\/section>\n
                \u4e2d\u5c71\u5927\u5b66\u6570\u5b66\u5b66\u9662\u6559\u6388\uff0c\u66fe\u662fUMass Amherst\u751f\u7269\u7edf\u8ba1\u52a9\u7406\u6559\u6388\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                https:\/\/zhichaoj-git.github.io\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Ben-Michael, E., Imai, K., & Jiang, Z. (2022). Policy learning with asymmetric utilities<\/em>.<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u5373\u4f7f\u5728\u533b\u5b66\u548c\u516c\u5171\u653f\u7b56\u7b49\u9ad8\u98ce\u9669\u73af\u5883\u4e2d\uff0c\u6570\u636e\u9a71\u52a8\u7684\u51b3\u7b56\u4e5f\u53d1\u6325\u7740\u91cd\u8981\u4f5c\u7528\u3002\u4ece\u89c2\u5bdf\u5230\u7684\u6570\u636e\u4e2d\u5b66\u4e60\u6700\u4f18\u7b56\u7565\u9700\u8981\u4ed4\u7ec6\u5730\u5236\u5b9a\u6548\u7528\u51fd\u6570\uff0c\u5176\u671f\u671b\u503c\u5728\u603b\u4f53\u4e2d\u6700\u5927\u5316\u3002\u867d\u7136\u7814\u7a76\u4eba\u5458\u901a\u5e38\u53ea\u4f7f\u7528\u4f9d\u8d56\u4e8e\u89c2\u5bdf\u7ed3\u679c\u7684\u6548\u7528\uff0c\u4f46\u5728\u8bb8\u591a\u60c5\u51b5\u4e0b\uff0c\u51b3\u7b56\u8005\u7684\u6548\u7528\u51fd\u6570\u66f4\u6070\u5f53\u5730\u8868\u5f81\u4e3a\u6240\u6709\u884c\u52a8\u4e0b\u6f5c\u5728\u7ed3\u679c\u7684\u8054\u5408\u96c6\u3002\u4f8b\u5982\uff0c\u5e0c\u6ce2\u514b\u62c9\u5e95\u7684\u201c\u4e0d\u4f24\u5bb3\u201d\u539f\u5219\u610f\u5473\u7740\uff0c\u5bf9\u4e00\u4e2a\u672c\u6765\u53ef\u4ee5\u4e0d\u63a5\u53d7\u6cbb\u7597\u800c\u5b58\u6d3b\u4e0b\u6765\u7684\u75c5\u4eba\u9020\u6210\u6b7b\u4ea1\u7684\u4ee3\u4ef7\u8981\u5927\u4e8e\u653e\u5f03\u633d\u6551\u751f\u547d\u7684\u6cbb\u7597\u7684\u4ee3\u4ef7\u3002\u6211\u4eec\u8003\u8651\u8fd9\u79cd\u5f62\u5f0f\u7684\u975e\u5bf9\u79f0\u6548\u7528\u51fd\u6570\u7684\u6700\u4f18\u7b56\u7565\u5b66\u4e60\u3002\u6211\u4eec\u8868\u660e\uff0c\u4e0d\u5bf9\u79f0\u7684\u6548\u7528\u5bfc\u81f4\u4e86\u4e00\u4e2a\u65e0\u6cd5\u8bc6\u522b\u7684\u793e\u4f1a\u798f\u5229\u51fd\u6570\uff0c\u56e0\u6b64\u6211\u4eec\u9996\u5148\u90e8\u5206\u5730\u8bc6\u522b\u4e86\u5b83\u3002\u5229\u7528\u7edf\u8ba1\u51b3\u7b56\u7406\u8bba\uff0c\u6211\u4eec\u901a\u8fc7\u6700\u5c0f\u5316\u76f8\u5bf9\u4e8e\u5907\u9009\u7b56\u7565\u7684\u6700\u5927\u540e\u6094\u6765\u63a8\u5bfc\u51fa\u6700\u5c0f\u51b3\u7b56\u89c4\u5219\u3002\u6211\u4eec\u8bc1\u660e\u4e86\u53ef\u4ee5\u4ece\u89c2\u5bdf\u5230\u7684\u6570\u636e\u5e93\u4e2d\u5b66\u4e60\u6781\u5927\u6781\u5c0f\u51b3\u7b56\u89c4\u5219\u6765\u89e3\u51b3\u4e2d\u95f4\u5206\u7c7b\u95ee\u9898\u3002\u6211\u4eec\u8fd8\u5efa\u7acb\u4e86\u8be5\u8fc7\u7a0b\u7684\u6709\u9650\u6837\u672c\u9057\u61be\u662f\u7531\u8fd9\u4e9b\u4e2d\u95f4\u5206\u7c7b\u5668\u7684\u8bef\u5206\u7c7b\u7387\u9650\u5236\u7684\u3002\u6211\u4eec\u5e94\u7528\u8fd9\u4e00\u6982\u5ff5\u6846\u67b6\u548c\u65b9\u6cd5\u6765\u51b3\u5b9a\u662f\u5426\u5bf9\u53ef\u80fd\u5b58\u5728\u80ba\u52a8\u8109\u9ad8\u538b\u7684\u60a3\u8005\u4f7f\u7528\u53f3\u5fc3\u5bfc\u7ba1\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u4f5c\u8005\u4f7f\u7528\u8fd9\u4e2a\u6846\u67b6\u91cd\u65b0\u8bc4\u4f30\u53f3\u5fc3\u5bfc\u7ba1 (RHC) \u7684\u4f7f\u7528\uff0c\u8fd9\u662f\u4e00\u79cd\u4fb5\u5165\u6027\u8bca\u65ad\u5de5\u5177 (Connors et al., 1996)\u3002\u4f5c\u8005\u6839\u636e\u4e34\u5e8a\u53d8\u91cf\u5b66\u4e60\u51b3\u7b56\u89c4\u5219\uff0c\u56e0\u4e3a\u4f5c\u8005\u5728\u672a\u80fd\u9632\u6b62\u60a3\u8005\u6b7b\u4ea1\u548c\u901a\u8fc7 RHC \u5bfc\u81f4\u5176\u6548\u7528\u51fd\u6570\u7684\u4e0d\u5bf9\u79f0\u6027\u5404\u4e0d\u76f8\u540c\u3002\u8fd9\u4e9b\u51b3\u7b56\u89c4\u5219\u56e0\u6211\u4eec\u662f\u5426\u6700\u5c0f\u5316\u76f8\u5bf9\u4e8e\u6052\u5b9a\u7b56\u7565\uff08\u603b\u662f\u6216\u4ece\u672a\u4f7f\u7528 RHC\uff09\u7684\u6700\u574f\u60c5\u51b5\u9057\u61be\u6216\u6700\u4f73\u4f7f\u7528 RHC \u7684\u9884\u8a00\u673a\u7b56\u7565\u800c\u6709\u6240\u4e0d\u540c\u3002\u4f5c\u8005\u68c0\u67e5\u4e86\u6548\u7528\u51fd\u6570\u548c\u6bd4\u8f83\u5668\u7684\u9009\u62e9\u5982\u4f55\u5f71\u54cd\u5b66\u4e60\u5230\u7684\u51b3\u7b56\u89c4\u5219\uff0c\u57fa\u4e8e\u8fd9\u4e9b\u9009\u62e9\u53d1\u73b0\u4e86\u5b9e\u8d28\u6027\u7684\u53ef\u53d8\u6027\u3002\u6700\u540e\uff0c\u4f5c\u8005\u5c06\u8fd9\u4e9b\u53d1\u73b0\u8f6c\u5316\u4e3a\u7ea6\u675f\u5bf9\u79f0\u6548\u7528\u6700\u5927\u5316\u6846\u67b6\uff0c\u4f5c\u8005\u5bfb\u6c42\u627e\u5230\u4e00\u79cd\u7b97\u6cd5\u7b56\u7565\uff0c\u8be5\u7b56\u7565\u6839\u636e\u7b56\u7565\u5371\u5bb3\u6216\u672a\u80fd\u8282\u7701\u7684\u60a3\u8005\u7684\u6700\u574f\u60c5\u51b5\u6bd4\u4f8b\u7684\u9650\u5236\u6765\u6700\u5c0f\u5316\u9884\u671f\u6b7b\u4ea1\u7387\u3002\u4f5c\u8005\u5728\u8fd9\u4e9b\u76ee\u6807\u4e4b\u95f4\u5c55\u793a\u4e86\u6743\u8861\uff0c\u8868\u660e\u9884\u671f\u6b7b\u4ea1\u7387\u5c06\u968f\u7740\u66f4\u5b57\u7b26\u4e32\u7684\u7ea6\u675f\u800c\u4ea7\u751f\u3002<\/span><\/section>\n


                <\/h1>\n
                \u6574\u7406\u8005\uff1a<\/span>\u6768\u660e\u54f2<\/span><\/section>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u91cd\u52a0\u6743\u4e0bRCT\u6cdb\u5316\uff1a\u6709\u9650\u6837\u672c\u8bef\u5dee\u548c\u53d8\u91cf\u9009\u62e9<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aReweighting the RCT for Generalization: Finite Sample Error and Variable Selection<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aBenedicte Colnet<\/span><\/section>\n
                \u6cd5\u56fd Inria \u4e09\u5e74\u7ea7\u5728\u8bfb\u535a\u58eb\u3002\u5df2\u52a0\u5165\u6cd5\u56fd\u533b\u7597\u793e\u4f1a\u4fdd\u969c\u90e8\u95e8\uff0c\u5173\u6ce8\u836f\u7269\u548c\u533b\u7597\u8bbe\u5907\u5b9a\u4ef7\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                https:\/\/benedictecolnet.github.io\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Colnet, B., Josse, J., Varoquaux, G., & Scornet, E. (2022). Reweighting the RCT for generalization: finite sample analysis and variable selection<\/em>.<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/h3>\n

                \u968f\u673a\u5bf9\u7167\u8bd5\u9a8c(RCT)\u7684\u6709\u9650\u8303\u56f4\u8d8a\u6765\u8d8a\u53d7\u5230\u5ba1\u67e5\uff0c\u7279\u522b\u662f\u5f53\u6837\u672c\u4e0d\u5177\u6709\u4ee3\u8868\u6027\u65f6\u3002\u4e8b\u5b9e\u4e0a\uff0c\u4e0e\u76ee\u6807\u4eba\u7fa4\u76f8\u6bd4\uff0c\u4e00\u4e9brct\u8d85\u8fc7\u6216\u4f4e\u4e8e\u6837\u672c\u7684\u4e2a\u4f53\u5177\u6709\u67d0\u4e9b\u7279\u5f81\uff0c\u56e0\u6b64\u4eba\u4eec\u60f3\u8981\u5f97\u51fa\u6cbb\u7597\u6709\u6548\u6027\u7684\u7ed3\u8bba\u3002\u91cd\u65b0\u52a0\u6743\u8bd5\u9a8c\u4e2a\u4f53\u4ee5\u5339\u914d\u76ee\u6807\u4eba\u7fa4\u6709\u52a9\u4e8e\u6539\u5584\u6cbb\u7597\u6548\u679c\u7684\u4f30\u8ba1\u3002\u8fd9\u6837\u7684\u7a0b\u5e8f\u9700\u8981\u4f30\u8ba1\u4e24\u4e2a\u5bc6\u5ea6\u7684\u6bd4\u7387(\u8bd5\u9a8c\u5206\u5e03\u548c\u76ee\u6807\u5206\u5e03)\u3002\u5728\u8fd9\u9879\u5de5\u4f5c\u4e2d\uff0c\u6211\u4eec\u5efa\u7acb\u4e86\u8fd9\u79cd\u91cd\u52a0\u6743\u8fc7\u7a0b\u7684\u504f\u5dee\u548c\u65b9\u5dee\u7684\u7cbe\u786e\u8868\u8fbe\u5f0f\u2014\u2014\u4e5f\u79f0\u4e3a\u62bd\u6837\u52a0\u6743\u9006\u503e\u5411(IPSW)\u2014\u2014\u5728\u4efb\u4f55\u6837\u672c\u91cf\u5b58\u5728\u5206\u7c7b\u534f\u53d8\u91cf\u7684\u60c5\u51b5\u4e0b\u3002\u8fd9\u6837\u7684\u7ed3\u679c\u4f7f\u6211\u4eec\u80fd\u591f\u6bd4\u8f83IPSW\u4f30\u8ba1\u7684\u4e0d\u540c\u7248\u672c\u7684\u7406\u8bba\u6027\u80fd\u3002\u6b64\u5916\uff0c\u6211\u4eec\u7684\u7ed3\u679c\u663e\u793aIPSW\u4f30\u8ba1\u7684\u6027\u80fd(\u504f\u5dee\uff0c\u65b9\u5dee\u548c\u4e8c\u6b21\u98ce\u9669)\u5982\u4f55\u53d6\u51b3\u4e8e\u4e24\u79cd\u6837\u672c\u91cf(RCT\u548c\u76ee\u6807\u4eba\u7fa4)\u3002\u6211\u4eec\u5de5\u4f5c\u7684\u4e00\u4e2a\u526f\u4ea7\u54c1\u662fIPSW\u4f30\u8ba1\u4e00\u81f4\u6027\u7684\u8bc1\u660e\u3002\u7cbe\u7ec6\u5316\u7684\u5206\u6790\u8fd8\u8868\u660e\uff0c\u5f53\u4f30\u8ba1\u5f85\u5904\u7406\u7684\u8bd5\u9a8c\u6982\u7387\u65f6(\u800c\u4e0d\u662f\u4f7f\u7528\u5bf9\u5e94\u7684oracle)\uff0c ipsw\u6027\u80fd\u4f1a\u5f97\u5230\u6539\u5584\u3002\u6b64\u5916\uff0c\u6211\u4eec\u8fd8\u7814\u7a76\u4e86\u5305\u62ec\u5bf9\u6b63\u786e\u4f30\u8ba1\u6743\u503c\u4e0d\u5fc5\u8981\u7684\u534f\u53d8\u91cf\u5982\u4f55\u5f71\u54cd\u6e10\u8fd1\u65b9\u5dee\u3002\u6211\u4eec\u7528\u4e24\u4e2a\u4f8b\u5b50\u6765\u8bf4\u660e\u6240\u6709\u7684\u7ed3\u8bba\uff1a\u4e00\u4e2a\u662f\u73a9\u5177\u548c\u8bf4\u6559\u7684\u4f8b\u5b50\uff0c\u53e6\u4e00\u4e2a\u662f\u53d7\u91cd\u75c7\u76d1\u62a4\u533b\u5b66\u542f\u53d1\u7684\u534a\u5408\u6210\u6a21\u62df\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u4f5c\u8005\u9996\u5148\u8bf4\u660e\u91cd\u65b0\u52a0\u6743\u7684\u539f\u5219\uff0c\u7279\u522b\u662f\uff0c\u5c55\u793a\u4e86 IPSW \u4f30\u8ba1\u5668\u7684\u51e0\u4e2a\u7248\u672c\uff1a\u8bd5\u9a8c\u6216\u76ee\u6807\u4eba\u7fa4\u7684\u534f\u53d8\u91cf\u6982\u7387\u662f\u5426\u4ece\u6570\u636e\u4e2d\u4f30\u8ba1\u6216\u5047\u8bbe\u4e3a\u9884\u8a00\u673a\u3002\u8fd9\u5c06\u4f5c\u8005\u7684\u7ed3\u679c\u4e0e\u56e0\u679c\u63a8\u7406\u548c\u6d41\u884c\u75c5\u5b66\u7684\u7ecf\u5178\u5de5\u4f5c\u8054\u7cfb\u8d77\u6765\u3002\u4f5c\u8005\u8fd8\u8be6\u7ec6\u8bf4\u660e\u4e86\u4e3a\u4ec0\u4e48\u53e6\u4e00\u4e2a\u7248\u672c\u7684IPSW\uff0c\u5176\u4e2d\u8fd8\u4f30\u8ba1\u4e86\u8bd5\u9a8c\u4e2d\u6cbb\u7597\u5206\u914d\u7684\u6982\u7387\uff0c\u65b9\u5dee\u8f83\u4f4e\u3002\u6700\u540e\u8ba8\u8bba\u4e86\u9644\u52a0\u548c\u975e\u5fc5\u8981\u534f\u53d8\u91cf\u5982\u4f55\u6539\u8fdb\u6216\u635f\u5bb3\u65b9\u5dee\uff0c\u5177\u4f53\u53d6\u51b3\u4e8e\u5b83\u4eec\u7684\u72b6\u6001\uff1a\u5b83\u4eec\u662f\u5426\u4ec5\u5728\u4e24\u4e2a\u79cd\u7fa4\u4e4b\u95f4\u79fb\u52a8\u6216\u4ec5\u5904\u7406\u6548\u5e94\u4fee\u9970\u7b26\u3002<\/span><\/section>\n


                <\/h1>\n
                \u6574\u7406\u8005\uff1a\u6768\u660e\u54f2<\/span><\/section>\n

                <\/span><\/section>\n

                <\/section>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u57fa\u4e8e\u5e73\u884c\u7ed3\u679c\u4fe1\u606f\u7684\u56e0\u679c\u63a8\u65ad<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aThe Promises of Parallel Outcomes<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1alinbo Wang<\/span><\/section>\n
                \u591a\u4f26\u591a\u5927\u5b66\u7edf\u8ba1\u79d1\u5b66\u7cfb\u548c\u8ba1\u7b97\u673a\u4e0e\u6570\u5b66\u79d1\u5b66\u7cfb\u52a9\u7406\u6559\u6388\u3002\u540c\u65f6\u4e5f\u662f\u77e2\u91cf\u7814\u7a76\u6240\u7684\u6559\u5e08\u3002\u6b64\u5916\uff0c\u4ed6\u8fd8\u62c5\u4efbCANSSI\u5b89\u5927\u7565STAGE\u9879\u76ee\u7684\u5bfc\u5e08\uff0c\u5e76\u62c5\u4efb\u534e\u76db\u987f\u5927\u5b66\u7edf\u8ba1\u7cfb\u548c\u591a\u4f26\u591a\u5927\u5b66\u8ba1\u7b97\u673a\u79d1\u5b66\u7cfb\u7684\u9644\u5c5e\u52a9\u7406\u6559\u6388\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875: https:\/\/sites.google.com\/site\/linbowangpku\/home<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Zhou, Y., Tang, D., Kong, D., & Wang, L. (2020). The Promises of Parallel Outcomes. arXiv: Methodology<\/em>.<\/span><\/section>\n


                <\/span><\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u5728\u8fdb\u884c\u89c2\u5bdf\u6027\u7814\u7a76\u65f6\uff0c\u8bc6\u522b\u548c\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u9762\u4e34\u7740\u4e00\u4e2a\u91cd\u8981\u6311\u6218\uff0c\u5373\u5982\u4f55\u5904\u7406\u5b58\u5728\u672a\u6d4b\u91cf\u6df7\u6742\u56e0\u7d20\u7684\u60c5\u51b5\u3002\u7814\u7a76\u4eba\u5458\u4ecb\u7ecd\u4e86\u4e00\u79cd\u65b0\u7684\u6846\u67b6\uff0c\u8be5\u6846\u67b6\u5229\u7528\u591a\u4e2a\u5e73\u884c\u7ed3\u679c\u4e2d\u7684\u4fe1\u606f\u6765\u8bc6\u522b\u548c\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u3002\u5728\u591a\u4e2a\u5e73\u884c\u7ed3\u679c\u4e4b\u95f4\u7684\u6761\u4ef6\u72ec\u7acb\u7ed3\u6784\u4e0b\uff0c\u5b9e\u73b0\u4e86\u81f3\u5c11\u4e09\u4e2a\u5e73\u884c\u7ed3\u679c\u7684\u975e\u53c2\u6570\u8bc6\u522b\u3002\u8fdb\u4e00\u6b65\u8868\u660e\uff0c\u5728\u4e00\u7ec4\u7ebf\u6027\u7ed3\u6784\u65b9\u7a0b\u6a21\u578b\u4e0b\uff0c\u56e0\u679c\u63a8\u7406\u662f\u53ef\u80fd\u7684\uff0c\u6709\u4e24\u4e2a\u5e73\u884c\u7684\u7ed3\u679c\u3002\u540c\u65f6\u5f00\u53d1\u4e86\u4f34\u968f\u7684\u4f30\u8ba1\u7a0b\u5e8f\uff0c\u5e76\u901a\u8fc7\u6a21\u62df\u7814\u7a76\u548c\u7814\u7a76tau\u86cb\u767d\u6c34\u5e73\u5bf9\u5404\u79cd\u884c\u4e3a\u7f3a\u9677\u7684\u56e0\u679c\u5f71\u54cd\u7684\u6570\u636e\u5e94\u7528\u6765\u8bc4\u4f30\u5176\u6709\u9650\u7684\u6837\u54c1\u6027\u80fd\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u5728\u8fd9\u7bc7\u8bba\u6587\u4e2d\uff0c\u4ecb\u7ecd\u4e86\u4e00\u79cd\u65b0\u7684\u56e0\u679c\u63a8\u65ad\u65b9\u6cd5\uff0c\u8be5\u65b9\u6cd5\u5229\u7528\u591a\u4e2a\u7ed3\u679c\u4e2d\u7684\u4fe1\u606f\u6765\u5904\u7406\u672a\u6d4b\u5230\u7684\u6df7\u6742\u56e0\u7d20\u3002\u8be5\u65b9\u6cd5\u57fa\u4e8e\u5e76\u884c\u7ed3\u679c\u7684\u5047\u8bbe\uff0c\u5373\u591a\u4e2a\u7ed3\u679c\u53d7\u5230\u76f8\u540c\u7684\u672a\u6d4b\u5230\u56e0\u7d20\u7684\u5f71\u54cd\u3002\u8bba\u6587\u4e2d\u63d0\u51fa\u4e86\u4e00\u79cd\u57fa\u4e8e\u7ed3\u6784\u65b9\u7a0b\u6a21\u578b\u7684\u65b9\u6cd5\u6765\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\uff0c\u5e76\u5728\u4e00\u5b9a\u7a0b\u5ea6\u4e0a\u8fdb\u884c\u4e86\u6d4b\u8bd5\u548c\u653e\u677e\u3002\u8be5\u65b9\u6cd5\u7684\u5173\u952e\u601d\u60f3\u662f\u5c06\u672a\u6d4b\u5230\u7684\u6df7\u6742\u56e0\u7d20\u8868\u793a\u4e3a\u66dd\u5149\u53d8\u91cf\u7684\u7ebf\u6027\u8f6c\u6362\u52a0\u4e0a\u9ad8\u65af\u8bef\u5dee\u3002\u8be5\u65b9\u6cd5\u7684\u4f18\u52bf\u5728\u4e8e\u53ef\u4ee5\u5904\u7406\u591a\u4e2a\u5e76\u884c\u7ed3\u679c\u7684\u60c5\u51b5\uff0c\u5e76\u4e14\u53ef\u4ee5\u5e94\u7528\u4e8e\u5e7f\u6cdb\u7684\u5b9e\u9645\u73af\u5883\u4e2d\u3002\u8be5\u65b9\u6cd5\u7684\u63d0\u51fa\u4e3a\u56e0\u679c\u63a8\u65ad\u63d0\u4f9b\u4e86\u4e00\u79cd\u6709\u5e0c\u671b\u7684\u66ff\u4ee3\u65b9\u6cd5\uff0c\u5e76\u4e14\u4e0e\u57fa\u4e8e\u591a\u56e0\u7d20\u6846\u67b6\u7684\u65b9\u6cd5\u6709\u7740\u6982\u5ff5\u4e0a\u7684\u533a\u522b\u3002\u5e76\u4e14\u6709\u671b\u5206\u6790\u9ad8\u7ef4\u54cd\u5e94\u6570\u636e\uff0c\u53ef\u4ee5\u4f7f\u7528\u989d\u5916\u7684\u7ed3\u679c\u6765\u653e\u5bbd\u6761\u4ef6\u72ec\u7acb\u5047\u8bbe\u3002<\/span><\/section>\n

                <\/section>\n
                \u6574\u7406\u8005\uff1a\u6bdb\u8fce\u8363<\/span><\/section>\n


                <\/h1>\n

                <\/span><\/section>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u5229\u7528\u989d\u5916\u7684\u968f\u673a\u6027\uff1a<\/strong><\/p>\n

                \u53ef\u590d\u5236\u6027\uff0c\u7075\u6d3b\u6027\u548c\u56e0\u679c\u5173\u7cfb<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aHarnessing Extra Randomness: Replicability, Flexibility and Causality<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aRichard Guo<\/span><\/section>\n
                \u5251\u6865\u5927\u5b66\u7edf\u8ba1\u5b9e\u9a8c\u5ba4\u7684\u7814\u7a76\u52a9\u7406\uff0c\u7531Rajen Shah\u6559\u6388\u6307\u5bfc\u3002\u6b64\u524d\uff0c\u4ed6\u662f\u897f\u8499\u65af\u8ba1\u7b97\u7406\u8bba\u7814\u7a76\u6240 2022 \u5e74\u56e0\u679c\u5173\u7cfb\u8ba1\u5212\u7684\u7406\u67e5\u5fb7\u00b7\u5361\u666e\u7814\u7a76\u5458\u3002\u4ed6\u4e8e 2021 \u5e74\u5728\u6258\u9a6c\u65af\u00b7\u7406\u67e5\u68ee\u7684\u6307\u5bfc\u4e0b\u83b7\u5f97\u534e\u76db\u987f\u5927\u5b66\u7edf\u8ba1\u5b66\u535a\u58eb\u5b66\u4f4d\u3002\u4ed6\u7684\u7814\u7a76\u5174\u8da3\u5305\u62ec\u56fe\u6a21\u578b\u3001\u56e0\u679c\u63a8\u7406\u3001\u534a\u53c2\u6570\u65b9\u6cd5\u548c\u6570\u636e\u5206\u6790\u7684\u53ef\u590d\u5236\u6027\u3002\u90ed\u535a\u58eb\u5c06\u4e8e2024\u5e741\u6708\u5f00\u59cb\u62c5\u4efb\u534e\u76db\u987f\u5927\u5b66\u751f\u7269\u7edf\u8ba1\u5b66\u52a9\u7406\u6559\u6388\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875: https:\/\/unbiased.co.in\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Guo, F. & Shah, Rajen. (2023). Rank-transformed subsampling: inference for multiple data splitting and exchangeable p-values. 10.48550\/arXiv.2301.02739.<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u8bb8\u591a\u73b0\u4ee3\u7edf\u8ba1\u7a0b\u5e8f\u662f\u968f\u673a\u7684\uff0c\u56e0\u4e3a\u8f93\u51fa\u662f\u6570\u636e\u7684\u968f\u673a\u51fd\u6570\u3002\u4f8b\u5982\uff0c\u8bb8\u591a\u8fc7\u7a0b\u91c7\u7528\u6570\u636e\u62c6\u5206\uff0c\u5c06\u6570\u636e\u96c6\u968f\u673a\u5212\u5206\u4e3a\u4e0d\u76f8\u4ea4\u7684\u90e8\u5206\u4ee5\u7528\u4e8e\u4e0d\u540c\u7684\u76ee\u7684\u3002\u5c3d\u7ba1\u5b83\u4eec\u5177\u6709\u7075\u6d3b\u6027\u548c\u53d7\u6b22\u8fce\u7a0b\u5ea6\uff0c\u4f46\u6570\u636e\u62c6\u5206\u548c\u5176\u4ed6\u968f\u673a\u7a0b\u5e8f\u7684\u7ed3\u6784\u5177\u6709\u660e\u663e\u7684\u7f3a\u70b9\u3002\u9996\u5148\uff0c\u7531\u4e8e\u5f15\u5165\u4e86\u989d\u5916\u7684\u968f\u673a\u6027\uff0c\u5bf9\u540c\u4e00\u6570\u636e\u96c6\u7684\u4e24\u6b21\u5206\u6790\u53ef\u80fd\u4f1a\u5bfc\u81f4\u4e0d\u540c\u7684\u7ed3\u679c\u3002\u5176\u6b21\uff0c\u968f\u673a\u7a0b\u5e8f\u901a\u5e38\u4f1a\u5931\u53bb\u7edf\u8ba1\u529f\u6548\uff0c\u56e0\u4e3a\u6574\u4e2a\u6837\u672c\u6ca1\u6709\u5145\u5206\u5229\u7528\u3002<\/span><\/p>\n

                \u4e3a\u4e86\u89e3\u51b3\u8fd9\u4e9b\u7f3a\u70b9\uff0c\u5728\u672c\u6b21\u6f14\u8bb2\u4e2d\uff0c\u6211\u5c06\u7814\u7a76\u5982\u4f55\u6b63\u786e\u7ec4\u5408\u968f\u673a\u8fc7\u7a0b\u7684\u591a\u6b21\u5b9e\u73b0\uff08\u4f8b\u5982\u901a\u8fc7\u591a\u4e2a\u6570\u636e\u62c6\u5206\uff09\u7684\u7ed3\u679c\u3002\u6211\u5c06\u4ecb\u7ecd\u79e9\u53d8\u6362\u5b50\u62bd\u6837\u4f5c\u4e3a\u5728\u6700\u5c0f\u5047\u8bbe\u4e0b\u63d0\u4f9b\u7ec4\u5408\u7ed3\u679c\u7684\u5927\u6837\u672c\u63a8\u65ad\u7684\u901a\u7528\u65b9\u6cd5\u3002\u6211\u5c06\u7528\u4e09\u4e2a\u5e94\u7528\u6765\u8bf4\u660e\u8be5\u65b9\u6cd5\uff1a\uff081\uff09\u4f7f\u7528\u9ad8\u7ef4\u57fa\u56e0\u8868\u8fbe\u6570\u636e\u68c0\u6d4b\u764c\u75c7\u4e9a\u578b\u7684\u7a0b\u5e8f\uff0c\uff082\uff09\u6821\u51c6\u4f7f\u7528\u4ea4\u53c9\u62df\u5408\u201c\u53cc\u91cd\u673a\u5668\u5b66\u4e60\u201d\u4f30\u8ba1\u7684\u56e0\u679c\u6548\u5e94\u7684\u7f6e\u4fe1\u533a\u95f4\uff0c\u4ee5\u53ca\uff083\uff09\u5728\u987a\u5e8f\u968f\u673a\u8bd5\u9a8c\u4e2d\u6d4b\u8bd5\u6ca1\u6709\u76f4\u63a5\u5f71\u54cd\u7684\u5047\u8bbe\u3002\u5bf9\u4e8e\u8fd9\u4e9b\u95ee\u9898\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u80fd\u591f\u53bb\u968f\u673a\u5316\u5e76\u63d0\u9ad8\u529f\u7387\u6216\u8986\u76d6\u8303\u56f4\u3002\u6b64\u5916\uff0c\u4e0e\u73b0\u6709\u7684\u7ec4\u5408p\u503c\u7684\u65b9\u6cd5\u76f8\u6bd4\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u4eab\u6709\u6e10\u8fd1\u63a5\u8fd1\u6807\u79f0\u6c34\u5e73\u7684I\u578b\u8bef\u5dee\u63a7\u5236\u3002\u8fd9\u4e00\u65b0\u53d1\u5c55\u5f00\u8f9f\u4e86\u8bbe\u8ba1\u663e\u5f0f\u968f\u673a\u5316\u548c\u53bb\u968f\u673a\u5316\u7a0b\u5e8f\u7684\u53ef\u80fd\u6027\uff1a\u5f15\u5165\u989d\u5916\u7684\u968f\u673a\u6027\u4ee5\u4f7f\u95ee\u9898\u5728\u88ab\u5220\u9664\u4e4b\u524d\u66f4\u5bb9\u6613\u3002\u6211\u8fd8\u5c06\u8ba8\u8bba\u8be5\u65b9\u6cd5\u5728\u56e0\u679c\u63a8\u7406\u548c\u56e0\u679c\u53d1\u73b0\u4e2d\u7684\u66f4\u5e7f\u6cdb\u5e94\u7528\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u8fd9\u7bc7\u6587\u7ae0\u7684\u4e3b\u8981\u8d21\u732e\u662f\u63d0\u51fa\u4e86\u4e00\u79cd\u57fa\u4e8e\u805a\u5408\u7684\u5047\u8bbe\u68c0\u9a8c\u65b9\u6cd5\uff0c\u540d\u4e3a\u201cdip hunting\u201d\u3002\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u9ad8\u7ef4\u5355\u5cf0\u68c0\u9a8c\u65b9\u6cd5;\u5e76\u4ecb\u7ecd\u4e86\u4e00\u79cd\u7b80\u5355\u3001\u7075\u6d3b\u7684\u53c2\u6570\u56de\u5f52\u6a21\u578b(\u5982\u53c2\u6570\u5206\u4f4d\u6570\u56de\u5f52)\u4f18\u5ea6\u68c0\u9a8c\u65b9\u6cd5\u3002\u8be5\u65b9\u6cd5\u901a\u8fc7\u5c06\u591a\u4e2a\u5047\u8bbe\u68c0\u9a8c\u7684p\u503c\u8fdb\u884c\u805a\u5408\uff0c\u4ee5\u63d0\u9ad8\u68c0\u9a8c\u7684\u6548\u679c\u3002\u4e0e\u5176\u4ed6\u5e38\u7528\u7684\u5047\u8bbe\u68c0\u9a8c\u65b9\u6cd5\u76f8\u6bd4\uff0c\u201cdip hunting\u201d\u65b9\u6cd5\u5728\u4fdd\u6301\u6b63\u786e\u7684\u663e\u8457\u6027\u6c34\u5e73\u7684\u540c\u65f6\uff0c\u80fd\u591f\u63d0\u4f9b\u66f4\u9ad8\u7684\u7edf\u8ba1\u529f\u6548\u3002\u8be5\u65b9\u6cd5\u5728\u591a\u4e2a\u5e94\u7528\u9886\u57df\u8fdb\u884c\u4e86\u5b9e\u8bc1\u7814\u7a76\uff0c\u5e76\u5c55\u73b0\u4e86\u826f\u597d\u7684\u6027\u80fd\u3002\u8fd9\u79cd\u805a\u5408\u7684\u5047\u8bbe\u68c0\u9a8c\u65b9\u6cd5\u5728\u5047\u8bbe\u68c0\u9a8c\u9886\u57df\u5177\u6709\u521b\u65b0\u6027\uff0c\u53ef\u4ee5\u4e3a\u7814\u7a76\u4eba\u5458\u63d0\u4f9b\u4e00\u79cd\u66f4\u6709\u6548\u7684\u7edf\u8ba1\u63a8\u65ad\u5de5\u5177\u3002<\/span><\/section>\n

                <\/span><\/section>\n
                \u6574\u7406\u8005\uff1a\u6bdb\u8fce\u8363<\/span><\/section>\n


                <\/h1>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u901a\u8fc7\u8fed\u4ee3\u56fe\u5c55\u5f00\u8fdb\u884c\u6df7\u6742\u9009\u62e9<\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aConfounder Selection Via Graph Expansion<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aQingyuan Zhao<\/span><\/section>\n
                \u5251\u6865\u5927\u5b66\u7eaf\u6570\u5b66\u4e0e\u6570\u7406\u7edf\u8ba1\u7cfb\u7edf\u8ba1\u5b9e\u9a8c\u5ba4\u7684\u52a9\u7406\u6559\u6388\uff0c\u827e\u4f26\u56fe\u7075\u7814\u7a76\u6240\u7814\u7a76\u5458<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875: http:\/\/www.statslab.cam.ac.uk\/~qz280\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

                Guo, F.R., & Zhao, Q. (2023). Confounder selection via iterative graph expansion. https:\/\/arxiv.org\/abs\/2309.06053<\/span><\/p>\n

                <\/h2>\n

                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u6df7\u6dc6\u9009\u62e9\u662f\u89c2\u5bdf\u8bbe\u8ba1\u4e2d\u6700\u5173\u952e\u7684\u6b65\u9aa4\u4e4b\u4e00\uff0c\u5373\u9009\u62e9\u4e00\u7ec4\u534f\u53d8\u91cf\u6765\u63a7\u5236\u6240\u6709\u6cbb\u7597\u548c\u7ed3\u679c\u7ec4\u4e4b\u95f4\u7684\u6df7\u6dc6\u3002\u6b64\u524d\u7684\u65b9\u6cd5\uff0c\u4f8b\u5982Pearl\u8457\u540d\u7684\u540e\u95e8\u6807\u51c6\uff0c\u901a\u5e38\u9700\u8981\u5148\u6307\u5b9a\u56e0\u679c\u533a\u95f4\uff0c\u8fd9\u5728\u5b9e\u8df5\u4e2d\u5f80\u5f80\u5f88\u56f0\u96be\u3002\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u7528\u4e8e\u8054\u5408\u9009\u62e9\u7684\u4ea4\u4e92\u5f0f\u7a0b\u5e8f\uff0c\u65e0\u9700\u9884\u5148\u6307\u5b9a\u56fe\u6216\u89c2\u5bdf\u53d8\u91cf\u96c6\u3002\u8be5\u8fc7\u7a0b\u901a\u8fc7\u8fed\u4ee3\u5730\u6269\u5c55\u56e0\u679c\u56fe\u5e76\u5bfb\u627e\u65b0\u7684\u201c\u4e3b\u8981\u8c03\u6574\u96c6\u201d\u6765\u89e3\u51b3\u5e95\u5c42\u56e0\u679c\u56fe\u53ef\u80fd\u5b58\u5728\u7f3a\u9677\u7684\u95ee\u9898\u3002\u7ed3\u6784\u4fe1\u606f\u4ee5\u4e3b\u8981\u8c03\u6574\u96c6\u5f62\u5f0f\u5b58\u5728\uff0c\u5e76\u6301\u7eed\u8fdb\u884c\u76f4\u5230\u627e\u5230\u80fd\u591f\u63a7\u5236\u521d\u59cb\u503c\u6216\u786e\u5b9a\u4e0d\u5b58\u5728\u8fd9\u53d8\u91cf\u96c6\u4e3a\u6b62\u3002\u6211\u4eec\u8bc1\u660e\u4e86\u5982\u679c\u7528\u6237\u5728\u6bcf\u4e2a\u6b65\u9aa4\u4e2d\u6b63\u786e\u6307\u5b9a\u4e3b\u8c03\u6574\u96c6\uff0c\u5219\u6211\u4eec\u7684\u8fc7\u7a0b\u65e2\u5065\u5168\u53c8\u5b8c\u6574\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u672c\u6587\u7684\u521b\u65b0\u4e4b\u5904\u5728\u4e8e\u5f00\u53d1\u4e86\u4e00\u79cd\u7cfb\u7edf\u65b9\u6cd5\uff0c\u4f7f\u7528\u65e0\u73af\u6709\u5411\u6df7\u5408\u56fe\uff08ADMG\uff09\u8fdb\u884c\u56e0\u679c\u63a8\u7406\u4e2d\u7684\u6df7\u6dc6\u9009\u62e9\u3002\u8be5\u65b9\u6cd5\u6d89\u53ca\u679a\u4e3e\u4e3b\u8981\u8c03\u6574\u96c6\u5e76\u4f7f\u7528\u6bcf\u4e2a\u4e3b\u8981\u8c03\u6574\u96c6\u6269\u5c55\u56fe\u5f62\u3002\u8fd9\u79cd\u65b9\u6cd5\u5141\u8bb8\u5f81\u6c42\u6709\u5173\u5185\u751f\u6027\u7684\u9886\u57df\u77e5\u8bc6\uff0c\u5e76\u4e14\u4e0d\u9700\u8981\u5b8c\u5168\u4e86\u89e3\u56e0\u679c\u7ed3\u6784\u6216\u4e00\u7ec4\u7ed9\u5b9a\u7684\u89c2\u5bdf\u5230\u7684\u534f\u53d8\u91cf\u3002\u672c\u6587\u8fd8\u8ba8\u8bba\u4e86\u8be5\u65b9\u6cd5\u7684\u5408\u7406\u6027\u548c\u5b8c\u6574\u6027\uff0c\u5e76\u5f3a\u8c03\u4e86\u4f7f\u7528\u56fe\u8868\u6765\u6539\u8fdb\u89c2\u5bdf\u6027\u7814\u7a76\u8bbe\u8ba1\u7684\u6f5c\u529b\u3002\u901a\u8fc7\u4f7f\u7528ADMG\u56fe\u6765\u63cf\u8ff0\u53d8\u91cf\u4e4b\u95f4\u7684\u56e0\u679c\u7ed3\u6784\uff0c\u7814\u7a76\u4eba\u5458\u53ef\u4ee5\u901a\u8fc7\u8fed\u4ee3\u67e5\u8be2\u7279\u5b9a\u7ed3\u6784\u6765\u9009\u62e9\u6df7\u6dc6\u53d8\u91cf\u3002\u8fd9\u79cd\u65b9\u6cd5\u4e0d\u9700\u8981\u5b8c\u5168\u4e86\u89e3\u56e0\u679c\u7ed3\u6784\u6216\u7ed9\u5b9a\u7684\u89c2\u5bdf\u534f\u53d8\u91cf\u96c6\u5408\uff0c\u63d0\u4f9b\u4e86\u4e00\u79cd\u7cfb\u7edf\u7684\u65b9\u5f0f\u6765\u83b7\u53d6\u5173\u4e8e\u5185\u751f\u6027\u7684\u9886\u57df\u77e5\u8bc6\u3002\u6587\u7ae0\u8fd8\u8ba8\u8bba\u4e86\u56fe\u6269\u5c55\u7684\u4f18\u70b9\uff0c\u4ee5\u53ca\u5982\u4f55\u5229\u7528\u56fe\u6a21\u578b\u6539\u8fdb\u89c2\u5bdf\u6027\u7814\u7a76\u8bbe\u8ba1\u7684\u53ef\u80fd\u6027\u3002<\/span><\/section>\n

                <\/span><\/section>\n
                \u6574\u7406\u8005\uff1a\u6bdb\u8fce\u8363<\/span><\/section>\n


                <\/h1>\n


                <\/strong><\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u5173\u4e8e\u5e73\u5747\u6cbb\u7597\u6548\u679c\u7684\u56de\u5f52\u8c03\u6574\u63d2\u8865\u4f30\u8ba1\u91cf<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aOn Regression-adjusted Imputation Estimators of the Average Treatment Effect<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aZhexiao Lin<\/span><\/section>\n
                Zhexiao Lin <\/span>\u535a\u58eb\u76ee\u524d\u5c31\u8bfb\u4e8e\u7f8e\u56fd\u52a0\u5dde\u5927\u5b66\u4f2f\u514b\u5229\u5206\u6821\u7edf\u8ba1\u5b66\u7cfb\uff0c\u5e08\u4ece\u4e01\u9e4f\u3001Peter Bickel \u6559\u6388\u3002\u4ed6\u4e8e2022\u5e74\u5728\u534e\u76db\u987f\u5927\u5b66\u83b7\u5f97\u7855\u58eb\u5b66\u4f4d\uff0c\u5e76\u4e8e2020\u5e74\u5728\u6d59\u6c5f\u5927\u5b66\u5b8c\u6210\u5b66\u58eb\u5b66\u4f4d\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875: https:\/\/zhexiaolin.github.io\/<\/span><\/section>\n<\/blockquote>\n


                <\/span><\/h2>\n

                \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                Lin, Zhexiao & Han, Fang. (2022). On regression-adjusted imputation estimators of the average treatment effect. 10.48550\/arXiv.2212.05424.<\/span><\/section>\n


                <\/span><\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u4f7f\u7528\u4f30\u8ba1\u7684\u56de\u5f52\u51fd\u6570\u6765\u7b97\u7f3a\u5931\u7684\u6f5c\u5728\u7ed3\u679c\u662f\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u7684\u81ea\u7136\u60f3\u6cd5\u3002\u5728\u6587\u732e\u4e2d\uff0c\u7ed3\u5408\u5f52\u7b97\u548c\u56de\u5f52\u8c03\u6574\u7684\u4f30\u8ba1\u5668\u88ab\u8ba4\u53ef\u4e0e\u589e\u5e7f\u9006\u6982\u7387\u52a0\u6743\u76f8\u5ab2\u7f8e\u3002\u56e0\u6b64\uff0c\u4eba\u4eec\u957f\u671f\u4ee5\u6765\u4e00\u76f4\u63a8\u6d4b\uff0c\u8fd9\u79cd\u4f30\u8ba1\u5668\u5728\u907f\u514d\u76f4\u63a5\u6784\u6743\u91cd\u7684\u540c\u65f6\uff0c\u4e5f\u5177\u6709\u53cc\u91cd\u9c81\u68d2\u6027\u3002\u672c\u6587\u63a8\u5e7f\u4e86\u4f5c\u8005\u7684\u65e9\u671f\u7ed3\u679c\uff0c\u5c06\u8fd9\u4e00\u731c\u60f3\u5f62\u5f0f\u5316\uff0c\u5e76\u8868\u660e\u5927\u91cf\u7ecf\u8fc7\u56de\u5f52\u8c03\u6574\u7684\u5f52\u7b97\u8ba1\u5e73\u5747\u5904\u7406\u6548\u679c\u786e\u5b9e\u5177\u5907\u53cc\u91cd\u53ef\u9760\u5916\uff0c\u5728\u6b63\u786e\u6307\u5b9a\u5bc6\u5ea6\u6a21\u578b\u548c\u56de\u5f52\u6a21\u578b\u6761\u4ef6\u4e0b\uff0c\u4eec\u53ef\u4ee5\u8bc1\u660e\u662f\u534a\u53c2\u6570\u6709\u6548\u7684\u3002\u6211\u4eec\u6240\u6db5\u76d6\u63d2\u503c\u65b9\u6cd5\u4e2d\u503c\u6ce8\u610f\u7684\u4f8b\u5b50\u5305\u62ec\u6838\u5339\u914d\u3001(\u52a0\u6743)\u6700\u8fd1\u90bb\u5339\u914d\u3001\u5c40\u90e8\u7ebf\u6027\u5339\u914d\u548c(\u5b9e)\u968f\u673a\u68ee\u6797\u3002<\/span><\/p>\n

                \u521b\u65b0<\/span><\/strong><\/h3>\n
                \u56de\u5f52\u548c\u5339\u914d\u65b9\u6cd5\u8054\u7cfb\u4e86\u56e0\u679c\u63a8\u7406\u548c\u7edf\u8ba1\u4e0a\u586b\u8865\u7f3a\u5931\u503c\u6587\u732e\u4e2d\u5e38\u7528\u7684\u4f30\u7b97\u65b9\u6cd5\u3002\u8fd9\u4e9b\u65b9\u6cd5\u672c\u8d28\u4e0a\u5229\u7528\u6838\u5339\u914d\u3001\u5c40\u90e8\u7ebf\u6027\u5339\u914d\u3001\u968f\u673a\u68ee\u6797\u6216\u6700\u8fd1\u90bb\u5339\u914d\u6765\u4f30\u7b97\u7f3a\u5931\u7684\u6f5c\u5728\u7ed3\u679c\u3002\u5355\u72ec\u4f7f\u7528\u8fd9\u4e9b\u4f30\u7b97\u65b9\u6cd5\u53ef\u80fd\u6548\u7387\u4f4e\u4e0b\u6216\u7f3a\u4e4f\u51c6\u786e\u6027\u3002\u56de\u5f52\u8c03\u6574\u540e\u7684\u4f30\u7b97\u65b9\u6cd5\u7684\u7406\u8bba\u7ed3\u679c\u975e\u5e38\u6709\u9650\u3002\u90e8\u5206\u539f\u56e0\u5728\u4e8e\u8fd9\u4e9b\u65b9\u6cd5\u5b8c\u5168\u7531\u7ed3\u679c\u6a21\u578b\u9a71\u52a8\uff0c\u56e0\u6b64\u4e0d\u6e05\u695a\u54ea\u4e00\u90e8\u5206\u8d77\u7740\u503e\u5411\u5f97\u5206\u52a0\u6743\u7684\u4f5c\u7528\u3002\u672c\u6587\u65e8\u5728\u63d0\u4f9b\u4e00\u4e2a\u901a\u7528\u7406\u8bba\uff0c\u4ee5\u63ed\u793a\u56de\u5f52\u8c03\u6574\u540e\u7684\u4f30\u7b97\u65b9\u6cd5\u7684\u6548\u7387\u548c\u9c81\u68d2\u6027\u3002\u4f5c\u8005\u5173\u6ce8\u4e00\u7c7b\u79f0\u4e3a\u7ebf\u6027\u5e73\u6ed1\u7684\u975e\u53c2\u6570\u56de\u5f52\u65b9\u6cd5\uff0c\u5176\u4e2d\u5305\u62ec\u4e0a\u8ff0\u6240\u6709\u793a\u4f8b(\u6838\u5339\u914d\u3001\u5c40\u90e8\u7ebf\u6027\u5339\u914d\u3001\u6700\u8fd1\u90bb\u5339\u914d\u548c\u968f\u673a\u68ee\u6797)\u3002\u65b0\u7406\u8bba\u8868\u660e\uff1a(1)\u7ebf\u6027\u5e73\u6ed1\u53ef\u4ee5\u9690\u5f0f\u5730\u4ea7\u751f\u5bc6\u5ea6\u6bd4\u4f30\u8ba1\u91cf\uff1b(2)Rubin(1973b)\u548cAbadie and Imbens(2011)\u5f62\u5f0f\u7684\u56de\u5f52\u8c03\u6574\u7684\u8f93\u5165\u65b9\u6cd5\u6784\u6210AIPW\u4f30\u8ba1\u91cf\uff1b(3)\u53ea\u8981\u6b63\u786e\u6307\u5b9a\u5bc6\u5ea6\u6a21\u578b\u6216\u7ed3\u679c\u6a21\u578b\uff0c\u8fd9\u4e9b\u8f93\u5165\u65b9\u6cd5\u662f\u4e00\u81f4\u7684\uff1b(4)\u5982\u679c\u5bc6\u5ea6\u548c\u7ed3\u679c\u6a21\u578b\u90fd\u88ab\u6b63\u786e\u6307\u5b9a\uff0c\u5b83\u4eec\u8fdb\u4e00\u6b65\u6784\u6210ATE\u7684\u6e10\u8fd1\u6b63\u6001\u4f30\u8ba1\uff0c\u5176\u6e10\u8fd1\u65b9\u5dee\u8fbe\u5230\u534a\u53c2\u6570\u6548\u7387\u4e0b\u754c(Hahn\uff0c 1998)\uff0c\u56e0\u6b64\u662f\u534a\u53c2\u6570\u6709\u6548\u7684\uff1b(5)\u56de\u5f52\u8c03\u6574\u540e\u7684\u53cc\u673a\u5668\u5b66\u4e60\u7248\u672c\u2014\u2014\u90a3\u4e9b\u901a\u8fc7\u6837\u672c\u5206\u88c2\u548c\u4ea4\u53c9\u62df\u5408\u6765\u4f30\u8ba1\u8f93\u5165\u51fd\u6570\u548c\u4fee\u6b63\u504f\u5dee\u7684\u65b9\u6cd5\u53ef\u4ee5\u5728\u524a\u5f31\u67d0\u4e9b\u6761\u4ef6\u7684\u540c\u65f6\u83b7\u5f97(3)\u548c(4)\u4e2d\u7684\u6027\u8d28\u3002\u672c\u6587\u7814\u7a76\u7ed3\u679c\u4e3a\u4f7f\u7528\u56de\u5f52\u8c03\u6574\u7684\u4f30\u7b97\u65b9\u6cd5\u63d0\u4f9b\u4e86\u5fc5\u8981\u7684\u7406\u8bba\u652f\u6301\uff0c\u5e76\u5c06\u5176\u5efa\u7acb\u4e3a\u57fa\u4e8e\u6743\u91cd\u7684\u4f30\u7b97\u65b9\u6cd5\u7684\u6709\u7528\u66ff\u4ee3\u65b9\u6848\u3002<\/span><\/section>\n

                <\/section>\n
                \u6574\u7406\u8005\uff1a\u6bdb\u8fce\u8363<\/span><\/section>\n


                <\/h1>\n


                <\/span><\/p>\n

                \n
                \n
                \n
                \n
                \n

                <\/section>\n

                <\/section>\n
                \n
                <\/svg><\/section>\n<\/section>\n<\/section>\n
                \n

                \u9634\u6027\u8bd5\u9a8c\u8bbe\u8ba1\u4e0b\u7684\u75ab\u82d7\u6709\u6548\u6027\u4f30\u8ba1\uff1a<\/strong><\/p>\n

                \u6761\u4ef6\u53ef\u4ea4\u6362\u4e0b\u56e0\u679c\u63a8\u65ad\u7684\u53ef\u8bc6\u522b\u6027\u548c\u6548\u7387\u7406\u8bba<\/strong><\/p>\n<\/section>\n

                \n

                <\/section>\n

                <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                <\/p>\n

                \n
                \u4e3b\u9898\uff1aVaccine Effectiveness Estimation Under the Test-negative Design: Identifiability and Efficiency Theory for Causal Inference Under Conditional Exchangeability<\/span><\/section>\n
                \u4e3b\u8bb2\u4eba\uff1aCong Jiang<\/span><\/section>\n
                \u52a0\u62ff\u5927\u8499\u7279\u5229\u5c14\u5927\u5b66\u535a\u58eb\u540e\uff0c\u52a0\u62ff\u5927\u6ed1\u94c1\u5362\u5927\u5b66\u535a\u58eb\u3002\u4ed6\u7684\u4e3b\u8981\u7814\u7a76\u5174\u8da3\u96c6\u4e2d\u5728\u63a8\u52a8\u56e0\u679c\u63a8\u7406\u65b9\u6cd5\u8bba\u9886\u57df\u3002\u5177\u4f53\u800c\u8a00\uff0c\u4ed6\u81f4\u529b\u4e8e\u52a8\u6001\u6cbb\u7597\u673a\u5236(DTRs)\u7684\u65b9\u6cd5\u8bba\u5f00\u53d1\u53ca\u5176\u5728\u73b0\u5b9e\u4e16\u754c\u4e2d\u7684\u5b9e\u9645\u5e94\u7528\u3002\u76ee\u524d\uff0c\u4ed6\u7684\u7814\u7a76\u96c6\u4e2d\u5728\u56e0\u679c\u63a8\u7406\u9886\u57df\u3002<\/span><\/section>\n
                \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/cjiang.netlify.app\/en\/<\/span><\/section>\n<\/blockquote>\n


                <\/h2>\n

                \u5f15\u7528\u6587\u732e\uff1a<\/span><\/strong><\/span><\/h2>\n
                Jiang C, Talbot D, Carazo S, et al. TNDDR: Efficient and doubly robust estimation of COVID-19 vaccine effectiveness under the test-negative design[J]. arXiv preprint arXiv:2310.04578, 2023.<\/span><\/section>\n


                <\/h2>\n

                \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                \u6458\u8981<\/span><\/strong><\/h3>\n

                \u68c0\u6d4b\u9634\u6027\u8bbe\u8ba1(TND)\u901a\u5e38\u7528\u4e8e\u76d1\u6d4b\u5b63\u8282\u6027\u6d41\u611f\u75ab\u82d7\u7684\u6709\u6548\u6027\uff0c\u6700\u8fd1\u5df2\u6210\u4e3aCOVID-19\u75ab\u82d7\u76d1\u6d4b\u7684\u4e00\u90e8\u5206\u3002\u4e0e\u75c5\u4f8b\u5bf9\u7167\u7814\u7a76\u4e0d\u540c\uff0c\u5b83\u62db\u52df\u5177\u6709\u5171\u540c\u75c7\u72b6\u8868\u73b0\u7684\u53c2\u4e0e\u8005\uff0c\u5e76\u5bf9\u4ed6\u4eec\u8fdb\u884c\u76ee\u6807\u611f\u67d3\u6d4b\u8bd5\u3002\u9633\u6027\u6d4b\u8bd5\u88ab\u8ba4\u4e3a\u662f\u201c\u6848\u4f8b\u201d\uff0c\u9634\u6027\u6d4b\u8bd5\u88ab\u8ba4\u4e3a\u662f\u201c\u63a7\u5236\u201d\u3002\u4f20\u7edf\u4e0a\uff0cLogistic\u56de\u5f52\u6821\u6b63\u4e86\u5e72\u6270\u56e0\u5b50\u4ee5\u4f30\u8ba1TND\u75ab\u82d7\u7684\u6709\u6548\u6027\u3002\u7136\u800c\uff0c\u5982\u679c\u5b58\u5728\u5e72\u6270\u56e0\u5b50\u5bf9\u6548\u679c\u7684\u5f71\u54cd\uff0c\u5219\u53ef\u80fd\u4f1a\u4ea7\u751f\u504f\u5dee\u3002\u672c\u6587\u9996\u5148\u56de\u987e\u4e86\u8fb9\u9645\u98ce\u9669\u6bd4\u7684\u6cbb\u7597\u52a0\u6743\u4f30\u8ba1\u5668\u7684\u53cd\u6982\u7387\uff0c\u8be5\u4f30\u8ba1\u5668\u5728\u6548\u5e94\u4fee\u6b63\u4e0b\u6709\u6548\uff0c\u4f46\u9700\u8981\u5bf9\u63a5\u79cd\u6982\u7387\u8fdb\u884c\u53c2\u6570\u5efa\u6a21\u3002\u4e3a\u89e3\u51b3\u8fd9\u4e00\u9650\u5236\uff0c\u672c\u6587\u63d0\u51fa\u4e00\u79cd\u65b0\u7684\u53cc\u91cd\u9c81\u68d2\u548c\u6709\u6548\u7684\u8fb9\u9645\u98ce\u9669\u6bd4\u7387\u4f30\u8ba1\u5668\u3002\u4ece\u7406\u8bba\u4e0a\u548c\u7ecf\u9a8c\u4e0a\u8bc1\u660e\u4e86\u901a\u8fc7\u5bf9\u5e72\u6270\u51fd\u6570\u7684\u673a\u5668\u5b66\u4e60\u5b9e\u73b0\u7684\u53c2\u6570\u6536\u655b\u7387\u3002<\/span><\/p>\n

                \u521b\u65b0\uff1a<\/span><\/strong><\/h3>\n
                  \n
                1. \n

                  \u68c0\u6d4b\u9634\u6027\u8bbe\u8ba1\uff08Test Negative Design\uff09\uff1a\u62a5\u544a\u805a\u7126\u4e8e\u4f7f\u7528\u68c0\u6d4b\u9634\u6027\u8bbe\u8ba1\uff08TND\uff09\u6765\u4f30\u7b97\u75ab\u82d7\u6548\u679c\u3002TND\u662f\u4e00\u79cd\u521b\u65b0\u7684\u65b9\u6cd5\uff0c\u7528\u4e8e\u5728\u771f\u5b9e\u4e16\u754c\u60c5\u5883\u4e2d\u7814\u7a76\u75ab\u82d7\u7684\u6709\u6548\u6027\u3002<\/span><\/p>\n<\/li>\n

                2. \n

                  \u53cc\u91cd\u7a33\u5065\u6027\uff1a\u62a5\u544a\u5f3a\u8c03\u4e86\u53cc\u91cd\u7a33\u5065\u6027\u7684\u6982\u5ff5\uff0c\u5373\u6240\u63d0\u51fa\u7684\u65b9\u6cd5\u5728\u503e\u5411\u5f97\u5206\u6a21\u578b\u6216\u7ed3\u679c\u6a21\u578b\u4e2d\u7684\u4efb\u4e00\u4e00\u4e2a\u672a\u6b63\u786e\u89c4\u5b9a\u7684\u60c5\u51b5\u4e0b\u4ecd\u7136\u4fdd\u6301\u4e00\u81f4\u548c\u6709\u6548\u3002\u8fd9\u662f\u65b9\u6cd5\u8bba\u7684\u4e00\u4e2a\u5173\u952e\u65b9\u9762\u3002<\/span><\/p>\n<\/li>\n

                3. \n

                  \u9ad8\u6548\u5f71\u54cd\u51fd\u6570\uff1a\u62a5\u544a\u4ecb\u7ecd\u4e86\u9ad8\u6548\u5f71\u54cd\u51fd\u6570\u7684\u6982\u5ff5\uff0c\u7528\u4e8e\u4f30\u8ba1\u75ab\u82d7\u6548\u679c\u3002\u8fd9\u662f\u4e00\u79cd\u5904\u7406\u6df7\u6dc6\u548c\u504f\u5dee\u7684\u521b\u65b0\u7edf\u8ba1\u65b9\u6cd5\u3002<\/span><\/p>\n<\/li>\n

                4. \n

                  \u673a\u5668\u5b66\u4e60\u65b9\u6cd5\uff1a\u62a5\u544a\u8ba8\u8bba\u4e86\u5728\u4f30\u7b97\u5e72\u6270\u51fd\u6570\u65f6\u4f7f\u7528\u673a\u5668\u5b66\u4e60\u65b9\u6cd5\uff0c\u8fd9\u662f\u4e00\u79cd\u73b0\u4ee3\u548c\u5148\u8fdb\u7684\u6570\u636e\u5206\u6790\u65b9\u6cd5\u3002<\/span><\/p>\n<\/li>\n

                5. \n

                  \u5b9e\u9645\u6570\u636e\u5206\u6790\uff1a\u62a5\u544a\u63d0\u4f9b\u4e86\u5728Omicron B.1.5\u4e3b\u5bfc\u65f6\u671f\u5bf9\u533b\u62a4\u4eba\u5458\u8fdb\u884c\u7684\u771f\u5b9e\u6570\u636e\u5206\u6790\u3002\u5c06\u8fd9\u4e00\u65b9\u6cd5\u5e94\u7528\u4e8e\u771f\u5b9e\u4e16\u754c\u6570\u636e\u662f\u62a5\u544a\u7684\u4e00\u4e2a\u6709\u4ef7\u503c\u548c\u5b9e\u9645\u7684\u65b9\u9762\u3002<\/span><\/p>\n<\/li>\n

                6. \n
                  \u5e72\u9884\u548c\u6700\u4f73\u6cbb\u7597\u5206\u914d\uff1a\u62a5\u544a\u63d0\u5230\u4e86\u5728\u4f30\u7b97\u75ab\u82d7\u6709\u6548\u6027\u4e2d\u8003\u8651\u5e72\u6270\u548c\u5f00\u53d1\u6700\u4f73\u6cbb\u7597\u5206\u914d\u7b56\u7565\u3002\u53cd\u6620\u4e86\u5904\u7406\u590d\u6742\u771f\u5b9e\u4e16\u754c\u60c5\u5883\u7684\u5148\u8fdb\u65b9\u6cd5\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

                  <\/section>\n
                  \u6574\u7406\u8005\uff1a\u738b\u6587\u793c<\/span><\/section>\n


                  <\/h1>\n


                  <\/span><\/p>\n

                  \n
                  \n
                  \n
                  \n
                  \n

                  <\/section>\n

                  <\/section>\n
                  \n
                  <\/svg><\/section>\n<\/section>\n<\/section>\n
                  \n

                  \u591a\u6e90\u6570\u636e\u4e0b\u7684\u5b50\u7fa4\u4f53\u5904\u7406\u6548\u5e94\u8fc1\u79fb<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                  \n

                  <\/section>\n

                  <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                  <\/p>\n

                  \n
                  \u4e3b\u9898\uff1aTransporting Subgroup Treatment Effects under Multi-source Data<\/span><\/section>\n
                  \u4e3b\u8bb2\u4eba\uff1aGuanbo Wang<\/span><\/section>\n
                  \u54c8\u4f5b\u5927\u5b66\u7684\u535a\u58eb\u540e\u7814\u7a76\u5458\uff0c\u4ed6\u5728\u9ea6\u5409\u5c14\u5927\u5b66\u83b7\u5f97\u4e86\u751f\u7269\u7edf\u8ba1\u5b66\u535a\u58eb\u5b66\u4f4d\uff082022\uff09\uff0c\u4ed6\u66fe\u57282021-2022\u671f\u95f4\u62c5\u4efbBiogen & Roche\u7684Consultant\u5b9e\u4e60\u751f\u3002\u4ed6\u7684\u7814\u7a76\u9886\u57df\u5305\u62ec\u56e0\u679c\u63a8\u65ad\u7684\u53ef\u8fc1\u79fb\u6027\uff0c\u5305\u62ec\u63a2\u7d22\u5f02\u8d28\u6027treatment effect\u7b49\u3002<\/span><\/section>\n
                  \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                  https:\/\/www.guanbowang.info\/<\/span><\/section>\n<\/blockquote>\n


                  <\/h2>\n

                  \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                  \u6458\u8981<\/span><\/strong><\/h3>\n

                  \u672c\u5de5\u4f5c\u5173\u6ce8\u4e8e\u56e0\u679c\u6548\u5e94\u7684\u8fc1\u79fb\u6027\u548c\u6cdb\u5212\u6027\uff0c\u65e8\u5728\u4f30\u7b97\u4e0d\u540c\u8bd5\u9a8c\u548c\u7fa4\u4f53\u4e2d\u7684treatment\u6548\u679c\uff0c\u4e0d\u540cmeta analysis\u7684\u7814\u7a76\uff0c\u672c\u4e13\u6ce8\u4e8e\u786e\u4fdd\u7edf\u8ba1\u4e25\u8c28\u6027\u5e76\u5e26\u6709\u56e0\u679c\u5047\u8bbe\u3002\u5b83\u8be6\u7ec6\u9610\u8ff0\u4e86\u4e00\u79cd\u5168\u9762\u7684\u65b9\u6cd5\uff0c\u7b26\u5408\u56e0\u679c\u5047\u8bbe\u5305\u62ec\u5b50\u7fa4\u4f53\u5206\u6790\u3001\u4f7f\u7528\u4e00\u81f4\u6027\u3001\u53ef\u4ea4\u6362\u6027\u548c\u6b63\u503c\u7b49\u3001\u5b9a\u4e49\u5404\u79cd\u6a21\u578b\u7684\u5e72\u6270\u53c2\u6570\uff0c\u5e76\u4f7f\u7528\u975e\u53c2\u6570\u5f71\u54cd\u51fd\u6570\u8fdb\u884c\u4f30\u7b97\u3002\u66f4\u8fdb\u4e00\u6b65\uff0c\u8be6\u7ec6\u7684\u7edf\u8ba1\u5206\u6790\u4e3a\u4f30\u7b97\u5668\u7684\u504f\u5dee\u548c\u65b9\u5dee\u63d0\u4f9b\u4e86\u89c1\u89e3\uff0c\u8bc1\u5b9e\u5728\u67d0\u4e9b\u6761\u4ef6\u4e0b\uff0c\u5b83\u53ef\u4ee5\u662f\u534a\u53c2\u6570\u6709\u6548\u7684\u3002\u901a\u8fc7\u4eff\u771f\u6570\u636e\u548c\u5173\u4e8e\u7cbe\u795e\u5206\u88c2\u75c7\u836f\u7269\u7684empirical\u8bd5\u9a8c\uff0c\u8fdb\u4e00\u6b65\u8bc1\u5b9e\u4e86\u8be5\u65b9\u6cd5\u7684\u5b9e\u7528\u6027\u548c\u7a33\u5065\u6027\u3002<\/span><\/p>\n

                  \u521b<\/span><\/strong>\u65b0<\/span><\/h3>\n
                  \u8be5\u65b9\u6cd5\u7684\u4e00\u4e2a\u663e\u8457\u521b\u65b0\u70b9\u662f\u5c06\u7edf\u8ba1\u5efa\u6a21\u3001\u56e0\u679c\u63a8\u65ad\u548c\u673a\u5668\u5b66\u4e60\u65e0\u7f1d\u6574\u5408\uff0c\u4ee5\u5e94\u5bf9\u4f30\u7b97\u5404\u79cd\u6570\u636e\u6e90\u548c\u5b50\u7fa4\u4f53\u4e2d\u6cbb\u7597\u6548\u679c\u65f6\u9047\u5230\u7684\u591a\u65b9\u9762\u590d\u6742\u6027\u3002\u8be5\u65b9\u6cd5\u8bbe\u8ba1\u7a33\u5065\u7684\u7edf\u8ba1\u8fc7\u7a0b\uff0c\u786e\u4fdd\u5b83\u4eec\u4e0e\u6570\u636e\u5185\u90e8\u590d\u6742\u7684\u56e0\u679c\u63a8\u65ad\u5173\u7cfb\u5f88\u597d\u5730\u914d\u5408\u3002\u4f30\u503c\u65b9\u6cd5\u7684\u7406\u8bba\u5c5e\u6027\u901a\u8fc7\u6a21\u62df\u548c\u5b9e\u9645\u5e94\u7528\u5f97\u5230\u4e86\u5b9e\u8d28\u6027\u9a8c\u8bc1\uff0c\u5728\u4eff\u771f\u548c\u771f\u5b9e\u6570\u636e\u4e0a\u9a8c\u8bc1\u4e86\u65b9\u6cd5\u6709\u6548\u6027\u3002<\/span><\/section>\n

                  <\/section>\n
                  \u6574\u7406\u8005\uff1a\u51af\u5e06<\/span><\/section>\n


                  <\/h1>\n


                  <\/span><\/p>\n

                  \n
                  \n
                  \n
                  \n
                  \n

                  <\/section>\n

                  <\/section>\n
                  \n
                  <\/svg><\/section>\n<\/section>\n<\/section>\n
                  \n

                  \u56e0\u679c\u56fe\u4e2d\u5b66\u4e60\u5f02\u8d28\u6027\u548c\u975e\u4f2a\u56e0\u6027<\/span><\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                  \n

                  <\/section>\n

                  <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                  <\/p>\n

                  \n
                  \u4e3b\u9898\uff1aTowards Causal Revolution: On Learning Heterogeneity and Non-Spuriousness in Causal Graphs<\/span><\/section>\n
                  \u4e3b\u8bb2\u4eba\uff1aHengrui Cai<\/span><\/section>\n
                  University of California Irvine \u7edf\u8ba1\u5b66\u7cfb\u7684\u52a9\u7406\u6559\u6388\u3002\u5979\u5728\u5317\u5361\u7f57\u6765\u7eb3\u5dde\u7acb\u5927\u5b66\uff08NCSU\uff09\u83b7\u5f97\u4e86\u7edf\u8ba1\u5b66\u535a\u58eb\u5b66\u4f4d\uff0c\u7531Dr. Wenbin Lu\u548cDr. Rui Song\u5171\u540c\u6307\u5bfc\u3002\u5728\u6b64\u4e4b\u524d\uff0c\u5979\u4e8e2017\u5e747\u6708\u4ece\u6d59\u6c5f\u5927\u5b66\u83b7\u5f97\u7edf\u8ba1\u5b66\u5b66\u58eb\u5b66\u4f4d\u3002\u5979\u5bf9\u56e0\u679c\u63a8\u65ad\u3001\u5f3a\u5316\u5b66\u4e60\u3001\u56fe\u5f62\u6a21\u578b\u53ca\u5176\u4ea4\u53c9\u9886\u57df\u7684\u65b9\u6cd5\u548c\u7406\u8bba\u90fd\u6709\u5e7f\u6cdb\u7684\u7814\u7a76\u5174\u8da3\uff0c\u76ee\u7684\u662f\u4e3a\u5404\u79cd\u771f\u5b9e\u4e16\u754c\u95ee\u9898\u5efa\u7acb\u53ef\u9760\u3001\u5f3a\u5927\u4e14\u53ef\u89e3\u91ca\u7684\u89e3\u51b3\u65b9\u6848\u3002\u76ee\u524d\uff0c\u5979\u7684\u4e3b\u8981\u7814\u7a76\u5de5\u4f5c\u5305\u62ec\u4f7f\u7528\u590d\u6742\u6570\u636e\u8fdb\u884c\u4e2a\u4f53\u5316\u7684\u6700\u4f18\u51b3\u7b56\u5236\u5b9a\u3001\u5f3a\u5316\/\u6df1\u5ea6\u5b66\u4e60\u4e2d\u7684\u653f\u7b56\u8bc4\u4f30\uff0c\u4ee5\u53ca\u9ad8\u7ef4\u4e2a\u4f53\u8c03\u89e3\u5206\u6790\u7684\u56e0\u679c\u5173\u7cfb\u53d1\u73b0\uff0c\u8fd9\u4e9b\u76f4\u63a5\u53d7\u5230\u7cbe\u51c6\u533b\u5b66\u3001\u5b9a\u5236\u7ecf\u6d4e\u5b66\u3001\u4e2a\u6027\u5316\u8425\u9500\u3001\u73b0\u4ee3\u6d41\u884c\u75c5\u5b66\u7b49\u9886\u57df\u7684\u6fc0\u52b1\u3002<\/span><\/section>\n
                  \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                  https:\/\/hengruicai.github.io\/<\/span><\/section>\n<\/blockquote>\n


                  <\/h2>\n

                  \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                  Zhang, Wenbo, et al. “Towards Trustworthy Explanation: On Causal Rationalization.” ICML 2023<\/span><\/section>\n


                  <\/h2>\n

                  \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                  \u6458\u8981<\/span><\/strong><\/h3>\n

                  \u6df1\u5ea6\u5b66\u4e60\u65f6\u4ee3\uff0c\u968f\u7740NLP\u7684\u8fdb\u6b65\uff0c\u7406\u6027\u5316\uff08rationalization\uff09\u6210\u4e3a\u4e86\u4e00\u4e2a\u5173\u952e\u7684\u6280\u672f\uff0c\u8be5\u6280\u672f\u901a\u8fc7\u9009\u62e9\u8f93\u5165\u6587\u672c\u7684\u5b50\u96c6\u6765\u89e3\u91ca\u9884\u6d4b\u4e2d\u7684\u4e3b\u8981\u53d8\u5316\u3002\u7136\u800c\uff0c\u73b0\u6709\u57fa\u4e8e\u5173\u8054\u7684\u7406\u6027\u5316\u65b9\u6cd5\u5728\u4e24\u4e2a\u6216\u66f4\u591a\u7247\u6bb5\u9ad8\u5ea6\u76f8\u5173\u5e76\u5bf9\u9884\u6d4b\u51c6\u786e\u6027\u63d0\u4f9b\u7c7b\u4f3c\u8d21\u732e\u65f6\uff0c\u4e0d\u80fd\u8bc6\u522b\u975e\u7ed3\u6784\u5316\u6570\u636e\u80cc\u540e\u771f\u6b63\u7684\u7406\u7531\uff0c\u8fd9\u88ab\u79f0\u4e3a\u201c\u4f2a\u5173\u8054\u6027\u201d\u6216spuriousness\u3002\u4e3a\u4e86\u89e3\u51b3\u8fd9\u4e00\u9650\u5236\uff0c\u4f5c\u8005\u4ece\u56e0\u679c\u63a8\u65ad\u7684\u89d2\u5ea6\uff0c\u5c06\u4e24\u4e2a\u56e0\u679c\u8981\u7d20\uff0c\u5373\u975e\u4f2a\u5173\u8054\u6027\u548c\u6548\u7387\uff0c\u5f15\u5165\u5230\u7406\u6027\u5316\u4e2d\u3002\u6211\u4eec\u57fa\u4e8e\u65b0\u63d0\u51fa\u7684\u7406\u6027\u5316\u7ed3\u6784\u56e0\u679c\u6a21\u578b\uff0c\u6b63\u5f0f\u5b9a\u4e49\u4e86\u4e00\u7cfb\u5217\u7684\u56e0\u679c\u6982\u7387\uff0c\u5e76\u5c06\u5176\u7406\u8bba\u8bc6\u522b\u786e\u5b9a\u4e3a\u5b66\u4e60\u5fc5\u8981\u548c\u5145\u5206\u7406\u7531\u7684\u4e3b\u8981\u7ec4\u6210\u90e8\u5206\u3002\u4e0e\u6700\u5148\u8fdb\u7684\u65b9\u6cd5\u76f8\u6bd4\uff0c\u6211\u4eec\u63d0\u51fa\u7684\u56e0\u679c\u7406\u6027\u5316\u5728\u771f\u5b9e\u7684\u8bc4\u8bba\u548c\u533b\u7597\u6570\u636e\u96c6\u4e0a\u8fdb\u884c\u4e86\u5e7f\u6cdb\u7684\u5b9e\u9a8c\uff0c\u5e76\u8bc1\u660e\u4e86\u5176\u6709\u6548\u6027\u3002<\/span><\/p>\n

                  \u521b\u65b0<\/span><\/strong><\/h3>\n
                  \u6587\u7ae0\u9996\u6b21\u4ece\u56e0\u679c\u63a8\u65ad\u7684\u89d2\u5ea6\u8003\u8651\u4e86\u7406\u6027\u5316\u95ee\u9898\uff0c\u5c06\u975e\u4f2a\u5173\u8054\u6027\u548c\u6548\u7387\u5f15\u5165\u4e3a\u5176\u4e3b\u8981\u7684\u56e0\u679c\u8981\u7d20\u3002\u6b64\u5916\uff0c\u4f5c\u8005\u63d0\u51fa\u4e86\u4e00\u4e2a\u65b0\u7684\u7406\u6027\u5316\u7ed3\u6784\u56e0\u679c\u6a21\u578b\uff0c\u5e76\u57fa\u4e8e\u8be5\u6a21\u578b\u5b9a\u4e49\u4e86\u4e00\u7cfb\u5217\u7684\u56e0\u679c\u6982\u7387\u3002\u8be5\u65b9\u6cd5\u7684\u7406\u8bba\u8bc6\u522b\u88ab\u786e\u5b9a\u4e3a\u5b66\u4e60\u5fc5\u8981\u548c\u5145\u5206\u7406\u7531\u7684\u4e3b\u8981\u7ec4\u6210\u90e8\u5206\uff0c\u8fd9\u662f\u4e0e\u73b0\u6709\u65b9\u6cd5\u4e0d\u540c\u7684\u91cd\u8981\u521b\u65b0\u3002\u901a\u8fc7\u5b9e\u9a8c\uff0c\u8bc1\u660e\u4e86\u8be5\u65b9\u6cd5\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\u76f8\u5bf9\u4e8e\u5f53\u524d\u73b0\u6709\u65b9\u6848\u5177\u6709\u66f4\u597d\u7684\u6027\u80fd\u3002<\/span><\/section>\n

                  <\/section>\n
                  \u6574\u7406\u8005\uff1a\u51af\u5e06<\/span><\/section>\n


                  <\/h1>\n


                  <\/span><\/p>\n

                  \n
                  \n
                  \n
                  \n
                  \n

                  <\/section>\n

                  <\/section>\n
                  \n
                  <\/svg><\/section>\n<\/section>\n<\/section>\n
                  \n

                  \u53ef\u80fd\u8bef\u8bbe\u7684\u8fb9\u7f18\u7ed3\u6784Cox\u6a21\u578b\u4e0b\u7684<\/strong><\/p>\n

                  \u53cc\u91cd\u7a33\u5065\u63a8\u65ad<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                  \n

                  <\/section>\n

                  <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                  <\/p>\n

                  \n
                  \u4e3b\u9898\uff1a<\/span>Doubly Robust Inference Under Possibly Misspecified Marginal Structural Cox Model<\/span><\/section>\n
                  \u4e3b\u8bb2\u4eba\uff1aRonghui Xu<\/span><\/section>\n
                  Dr. Xu \u5728\u52a0\u5dde\u5927\u5b66\u5723\u5730\u4e9a\u54e5\u5206\u6821\uff08UCSD\uff09\u5206\u522b\u83b7\u5f97\u4e86\u5e94\u7528\u6570\u5b66\u7855\u58eb\uff081995\u5e74\uff09\u548c\u6570\u5b66\u535a\u58eb\uff081996\u5e74\uff09\u5b66\u4f4d\u3002\u5979\u5728UCSD\u764c\u75c7\u4e2d\u5fc3\u548c\u6570\u5b66\u7cfb\u62c5\u4efb\u4e86\u4e00\u5e74\u7684\u535a\u58eb\u540e\u7814\u7a76\u5458\uff0c\u7136\u540e\u57281997\u5e74\u5f00\u59cb\u5728\u54c8\u4f5b\u5927\u5b66\u516c\u5171\u536b\u751f\u5b66\u9662\u7684\u751f\u7269\u7edf\u8ba1\u5b66\u7cfb\u4ee5\u53ca\u6ce2\u58eb\u987f\u9a6c\u8428\u8bf8\u585e\u5ddeDana-Farber\u764c\u75c7\u7814\u7a76\u6240\u7684\u751f\u7269\u7edf\u8ba1\u79d1\u5b66\u7cfb\u62c5\u4efb\u52a9\u7406\u6559\u6388\u3002\u5979\u4e8e2004\u5e74\u56de\u5230UCSD\uff0c\u6210\u4e3a\u516c\u5171\u536b\u751f\u7cfb\u4ee5\u53ca\u6570\u5b66\u7cfb\u7684\u526f\u6559\u6388\u3002\u5979\u662fDavid P. Byar Young Investigator Award\u7684\u83b7\u5f97\u8005\uff0c\u5e76\u4e14\u662f\u7f8e\u56fd\u7edf\u8ba1\u534f\u4f1a\uff08American Statistical Association\uff09\u7684\u4f1a\u5458\u3002<\/span><\/section>\n
                  \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                  https:\/\/mathweb.ucsd.edu\/~rxu\/<\/span><\/section>\n<\/blockquote>\n


                  <\/span><\/h2>\n

                  \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

                  Hou, Jue, Jelena Bradic, and Ronghui Xu. “Treatment effect estimation under additive hazards models with high-dimensional confounding.” Journal of the American Statistical Association<\/em> 118.541 (2023): 327-342.<\/span><\/p>\n

                  Rava, Denise, and Ronghui Xu. “Doubly robust estimation of the hazard difference for competing risks data.” Statistics in Medicine 42.6 (2023): 799-814.<\/span><\/p>\n

                  Wang, Yuyao, Andrew Ying, and Ronghui Xu. “Doubly Robust Estimation under Covariate-induced Dependent Left Truncation.” arXiv preprint arXiv:2208.06836<\/em> (2022).<\/span><\/p>\n

                  Luo, Jiyu, and Ronghui Xu. “Doubly robust inference for hazard ratio under informative censoring with machine learning.” arXiv preprint arXiv:2206.02296 (2022).<\/span><\/p>\n

                  Zhang, Yiran, and Ronghui Xu. “Marginal Structural Illness-Death Models for Semi-Competing Risks Data.” arXiv preprint arXiv:2204.10426 (2022).<\/span><\/p>\n

                  Ying, Andrew, and Ronghui Xu. “On defense of the hazard ratio.” arXiv preprint arXiv:2307.11971 (2023).<\/span><\/p>\n

                  \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                  \u6458\u8981<\/span><\/strong><\/h3>\n

                  \u8be5\u5de5\u4f5c\u6df1\u5165\u63a2\u8ba8\u4e86\u751f\u5b58\u5206\u6790\u5728\u53ef\u80fd\u9519\u8bef\u6307\u5b9a\u7684\u8fb9\u7f18\u7ed3\u6784Cox\u6a21\u578b\u4e0b\u7684\u53cc\u91cd\u7a33\u5065\u63a8\u65ad\u65b9\u6cd5\u3002\u5de5\u4f5c\u6db5\u76d6\u4e86\u4e00\u7cfb\u5217\u590d\u6742\u7684\u7edf\u8ba1\u65b9\u6cd5\u8bba\uff0c\u5305\u62ec\u8be6\u7ec6\u7684\u751f\u5b58\u5206\u6790\u3001\u6f5c\u5728\u7ed3\u679c\u6846\u67b6\u7684\u5e94\u7528\u3001\u5371\u9669\u51fd\u6570\u53ca\u5176\u5728Cox\u6a21\u578b\u4e2d\u7684\u8fd0\u7528\u3001treatment effect\u7684\u6df1\u5ea6\u5206\u6790\u7b49\u3002\u901a\u8fc7\u5b9e\u9645\u7684\u6a21\u62df\u7ed3\u679c\u548c\u771f\u5b9e\u4e16\u754c\u7684\u5e94\u7528\u6848\u4f8b\uff0c\u5982\u706b\u5974\u9c81\u9c81\u7684\u4e00\u4e2a\u4e2d\u5e74\u996e\u9152\u8001\u5316\u7814\u7a76\uff0c\u5c55\u793a\u4e86\u5404\u79cd\u7edf\u8ba1\u65b9\u6cd5\u5728\u7406\u8bba\u548c\u5b9e\u8df5\u5e94\u7528\u4e2d\u7684\u7efc\u5408\u8fd0\u7528\u548c\u9a8c\u8bc1\uff0c\u7279\u522b\u7a81\u51fa\u4e86\u9006\u6982\u7387\u52a0\u6743\uff08IPW\uff09\u548c\u589e\u5f3a\u9006\u6982\u7387\u52a0\u6743\uff08AIPW\uff09\u4f30\u8ba1\u51fd\u6570\u5728\u63d0\u4f9b\u53ef\u80fd\u7684\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\u65b9\u9762\u7684\u521b\u65b0\u5b9e\u8df5\uff0c\u5e76\u5728\u5404\u79cd\u6a21\u62df\u573a\u666f\u4e0b\u9a8c\u8bc1\u4e86\u8fd9\u4e9b\u65b9\u6cd5\u7684\u7a33\u5065\u6027\u3002\u63d0\u4f9b\u4e86\u5173\u4e8e\u5982\u4f55\u5728\u53ef\u80fd\u7684\u6570\u636e\u751f\u6210\u8fc7\u7a0b\u7684\u4e0d\u540c\u771f\u5b9e\u6027\u4e0b\uff0c\u5229\u7528\u8fd9\u4e9b\u590d\u6742\u7684\u7edf\u8ba1\u65b9\u6cd5\u8fdb\u884c\u7a33\u5065\u7684\u56e0\u679c\u63a8\u65ad\u7684\u65b9\u6cd5\u3002<\/span><\/p>\n

                  \u521b\u65b0<\/span><\/strong><\/h3>\n

                  \u8be5\u7cfb\u5217\u5de5\u4f5c\u5728\u6f5c\u5728\u7ed3\u679c\u6846\u67b6\u548c\u5371\u9669\u51fd\u6570\u7684\u8ba8\u8bba\u4e2d\uff0c\u878d\u5408\u4e86\u4f20\u7edf\u7684Cox\u6a21\u578b\u4e0e\u6f5c\u5728\u7ed3\u679c\u6846\u67b6\uff0c\u5c24\u5176\u5728\u5904\u7406\u201cbeta t\u201d\uff08\u5bf9\u6570\u5371\u9669\u6bd4\u7387\uff09\u8fd9\u4e00\u53ef\u968f\u65f6\u95f4\u53d8\u5316\u7684\u53d8\u91cf\u65f6\u8868\u73b0\u51fa\u72ec\u5230\u89c1\u89e3\u3002\u5176\u6b21\uff0c\u5728\u5904\u7406\u7f3a\u5931\u6570\u636e\u7684\u95ee\u9898\u4e0a\uff0c\u6df1\u5165\u63a2\u8ba8\u4e86\u5305\u62ec\u4e00\u81f4\u6027\u3001\u65e0\u672a\u6d4b\u6df7\u6742\u548c\u91cd\u53e0\u5047\u8bbe\u5728\u5185\u7684\u591a\u79cd\u5047\u8bbe\u3002\u53e6\u5916\u4e00\u4e2a\u663e\u8457\u7684\u521b\u65b0\u70b9\u5728\u4e8e\u5229\u7528\u9006\u6982\u7387\u52a0\u6743\uff08IPW\uff09\u548c\u589e\u5f3a\u9006\u6982\u7387\u52a0\u6743\uff08AIPW\uff09\u4f30\u8ba1\u51fd\u6570\uff0c\u5c24\u5176\u662f\u901a\u8fc7\u5c06IPW\u4e0eAIPW\u4f30\u8ba1\u51fd\u6570\u76f8\u7ed3\u5408\uff0c\u63d0\u4f9b\u4e86\u4e00\u4e2a\u6f5c\u5728\u7684\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\uff0c\u589e\u5f3a\u4e86\u5728\u53ef\u80fd\u9519\u8bef\u6307\u5b9a\u7684\u6a21\u578b\u4e0b\u8fdb\u884c\u63a8\u65ad\u7684\u7a33\u5065\u6027\u3002\u5728\u6a21\u62df\u7ed3\u679c\u548c\u771f\u5b9e\u4e16\u754c\u5e94\u7528\u7684\u5b9e\u9a8c\u90e8\u5206\uff0c\u6709\u6548\u5730\u5229\u7528\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\u6765\u4f30\u8ba1\u5e72\u6270\u53c2\u6570\uff0c\u5e76\u5728\u5404\u79cd\u6a21\u62df\u573a\u666f\u4e0b\u5c55\u793a\u4e86\u591a\u79cd\u4f30\u8ba1\u5668\u7684\u6027\u80fd\uff0c\u8fd9\u540c\u6837\u4f53\u73b0\u4e86\u7406\u8bba\u4e0e\u5b9e\u9645\u5e94\u7528\u7684\u7d27\u5bc6\u7ed3\u5408\uff0c\u8868\u660e\u4e86\u8fd9\u4e00\u65b9\u6cd5\u8bba\u5728\u5b9e\u9645\u7814\u7a76\u4e2d\u7684\u5e94\u7528\u6f5c\u529b\u3002\u8fd9\u4e9b\u521b\u65b0\u70b9\u4e00\u65b9\u9762\u4f53\u73b0\u4e86\u5bf9\u4f20\u7edf\u65b9\u6cd5\u7684\u6df1\u5165\u7406\u89e3\u548c\u5e94\u7528\uff0c\u53e6\u4e00\u65b9\u9762\u5c55\u793a\u4e86\u5728\u590d\u6742\u6a21\u578b\u548c\u591a\u53d8\u5e94\u7528\u573a\u666f\u4e2d\u7a33\u5065\u4f30\u8ba1\u548c\u63a8\u65ad\u65b9\u6cd5\u7684\u63a2\u7d22\u548c\u53d1\u5c55\u3002<\/span><\/h3>\n

                  <\/section>\n
                  \u6574\u7406\u8005\uff1a\u51af\u5e06<\/span><\/section>\n


                  <\/h1>\n


                  <\/span><\/p>\n

                  \n
                  \n
                  \n
                  \n
                  \n

                  <\/section>\n

                  <\/section>\n
                  \n
                  <\/svg><\/section>\n<\/section>\n<\/section>\n
                  \n

                  \u5982\u4f55\u5728\u8bc4\u4f30\u6700\u4f18\u52a8\u6001\u6cbb\u7597\u65b9\u6848\u65f6<\/strong><\/p>\n

                  \u5229\u7528\u6f5c\u5728\u7684\u60a3\u8005\u504f\u597d\uff1f<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                  \n

                  <\/section>\n

                  <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                  <\/p>\n

                  \n
                  \u4e3b\u9898: How to Use Latent Patient Preference when Evaluating the Optimal Dynamic Treatment Regimes?<\/span><\/section>\n
                  \u4e3b\u8bb2\u4eba\uff1aLu Wang\uff0c\u5bc6\u6b47\u6839\u5927\u5b66\u6559\u6388<\/span><\/section>\n
                  Lu Wang \u535a\u58eb\u4e8e2008\u5e74\u5728\u54c8\u4f5b\u5927\u5b66\u83b7\u5f97\u751f\u7269\u7edf\u8ba1\u5b66\u535a\u58eb\u5b66\u4f4d\uff0c\u5e76\u4e8e\u540c\u5e74\u52a0\u5165\u5bc6\u6b47\u6839\u5927\u5b66\u7684\u6559\u804c\u3002\u5979\u7684\u7814\u7a76\u91cd\u70b9\u5305\u62ec\u8bc4\u4f30\u52a8\u6001\u6cbb\u7597\u65b9\u6848\u3001\u4e2a\u6027\u5316\u533b\u7597\u3001\u975e\u53c2\u6570\u548c\u534a\u53c2\u6570\u56de\u5f52\u3001\u7f3a\u5931\u6570\u636e\u5206\u6790\u3001\u51fd\u6570\u6570\u636e\u5206\u6790\u4ee5\u53ca\u7eb5\u5411\uff08\u76f8\u5173\/\u96c6\u7fa4\uff09\u6570\u636e\u5206\u6790\u7684\u7edf\u8ba1\u65b9\u6cd5\u3002\u738b\u535a\u58eb\u4e0e\u5b89\u5fb7\u68ee\u764c\u75c7\u4e2d\u5fc3\u3001\u5bc6\u6b47\u6839\u5927\u5b66\u533b\u5b66\u9662\u3001\u5bc6\u6b47\u6839\u5927\u5b66\u516c\u5171\u536b\u751f\u5b66\u9662\u3001\u5bc6\u6b47\u6839\u5927\u5b66\u8fd0\u52a8\u5b66\u9662\u3001\u5bc6\u6b47\u6839\u5927\u5b66\u793e\u4f1a\u7814\u7a76\u5b66\u9662\u548c\u54c8\u4f5b\u516c\u5171\u536b\u751f\u5b66\u9662\u7684\u7814\u7a76\u4eba\u5458\u5408\u4f5c\u3002\u738b\u535a\u58eb\u662f\u300a\u7f8e\u56fd\u7edf\u8ba1\u5b66\u4f1a\u6742\u5fd7\u300b\u548c\u300a\u751f\u7269\u7edf\u8ba1\u5b66\u300b\u7684\u526f\u7f16\u8f91\u3002\u5979\u8fd8\u62c5\u4efb\u5bc6\u6b47\u6839\u5927\u5b66\u751f\u547d\u5468\u671f\u73af\u5883\u66b4\u9732\u4e0e\u75be\u75c5\u7efc\u5408\u5065\u5eb7\u79d1\u5b66\u6838\u5fc3\uff08IHSC\uff09\u7684\u6838\u5fc3\u5408\u4f5c\u7ec4\u957f\u3002<\/span><\/section>\n<\/blockquote>\n


                  <\/h2>\n

                  \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

                  Zhong, Yingchao, Chang Wang, and Lu Wang. “Survival Augmented Patient Preference Incorporated Reinforcement Learning to Evaluate Tailoring Variables for Personalized Healthcare.” Stats<\/em> 4.4 (2021): 776-792.<\/span><\/p>\n

                  \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                  \u6458\u8981<\/span><\/strong><\/h3>\n

                  \u672c\u6587\u8003\u8651\u4e2a\u6027\u5316\u6cbb\u7597\u51b3\u7b56\u7b56\u7565\u5728\u6162\u6027\u75be\u75c5\u7ba1\u7406\u4e2d\u7684\u5e94\u7528\uff0c\u4f8b\u5982\u6162\u6027\u80be\u810f\u75c5\uff0c\u5176\u901a\u5e38\u5305\u62ec\u987a\u5e8f\u548c\u81ea\u9002\u5e94\u7684\u6cbb\u7597\u51b3\u7b56\u3002\u6211\u4eec\u7814\u7a76\u4e86\u4e00\u4e2a\u5177\u6709\u53ef\u80fd\u88ab\u53f3\u622a\u5c3e\u7684\u751f\u5b58\u7ed3\u679c\u7684\u4e24\u9636\u6bb5\u6cbb\u7597\u8bbe\u7f6e\u3002\u8fd9\u53ef\u4ee5\u901a\u8fc7\u52a8\u6001\u6cbb\u7597\u89c4\u5219\uff08DTR\uff09\u6846\u67b6\u6765\u8fdb\u884c\u5efa\u6a21\uff0c\u5176\u76ee\u6807\u662f\u6839\u636e\u6bcf\u4e2a\u4e2a\u4f53\u7684\u533b\u7597\u5386\u53f2\u6765\u4e2a\u6027\u5316\u6cbb\u7597\uff0c\u4ee5\u6700\u5927\u5316\u7406\u60f3\u7684\u5065\u5eb7\u7ed3\u679c\u3002\u6211\u4eec\u5f00\u53d1\u4e86\u4e00\u79cd\u65b0\u7684\u65b9\u6cd5\uff0c\u79f0\u4e3aSurvival Augmented Patient Preference incorporated reinforcement Q-Learning\uff08SAPP-Q-Learning\uff09\uff0c\u7528\u4e8e\u5728\u6700\u5927\u968f\u8bbf\u671f\u9650\u5185\u5728\u751f\u6d3b\u8d28\u91cf\u548c\u751f\u5b58\u4e4b\u95f4\u8fdb\u884c\u51b3\u7b56\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u5c06\u60a3\u8005\u504f\u597d\u7eb3\u5165\u52a0\u6743\u6548\u7528\u51fd\u6570\u4e2d\uff0c\u5e73\u8861\u751f\u6d3b\u8d28\u91cf\u548c\u751f\u5b58\u65f6\u95f4\uff0c\u5728Q-learning\u6a21\u578b\u6846\u67b6\u4e2d\u8fdb\u884c\u5efa\u6a21\u3002\u6211\u4eec\u8fdb\u4e00\u6b65\u63d0\u51fa\u4e86\u76f8\u5e94\u7684m-out-of-n Bootstrap\u8fc7\u7a0b\uff0c\u4ee5\u51c6\u786e\u8fdb\u884c\u7edf\u8ba1\u63a8\u65ad\uff0c\u5e76\u6784\u5efa\u5173\u4e8e\u4e2a\u6027\u5316\u6cbb\u7597\u7b56\u7565\u6307\u5bfc\u53d8\u91cf\u6548\u679c\u7684\u7f6e\u4fe1\u533a\u95f4\u3002<\/span><\/p>\n

                  \u521b\u65b0<\/span><\/strong><\/h3>\n
                    \n
                  1. \n

                    \u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u65b9\u6cd5\uff0c\u79f0\u4e3aSurvival Augmented Patient Preference incorporated reinforcement Q-Learning (SAPP-Q-Learning)\uff0c\u7528\u4e8e\u5728\u4e2a\u6027\u5316\u533b\u7597\u4e2d\u8bc4\u4f30\u6cbb\u7597\u53d8\u91cf\u3002\u8be5\u65b9\u6cd5\u7ed3\u5408\u4e86\u60a3\u8005\u7684\u6f5c\u5728\u504f\u597d\u4e0e\u751f\u5b58\u65f6\u95f4\u4e4b\u95f4\u7684\u6743\u8861\uff0c\u901a\u8fc7\u52a0\u6743\u6548\u7528\u51fd\u6570\u5728Q-Learning\u6a21\u578b\u6846\u67b6\u4e2d\u8fdb\u884c\u51b3\u7b56\uff0c\u4ee5\u5728\u6700\u5927\u968f\u8bbf\u671f\u9650\u4e0b\u5e73\u8861\u751f\u6d3b\u8d28\u91cf\u548c\u751f\u5b58\u65f6\u95f4\u3002<\/span><\/p>\n<\/li>\n

                  2. \n

                    \u63d0\u51fa\u4e86\u4e00\u79cd\u76f8\u5e94\u7684m-out-of-n Bootstrap\u7a0b\u5e8f\uff0c\u7528\u4e8e\u51c6\u786e\u8fdb\u884c\u7edf\u8ba1\u63a8\u65ad\uff0c\u5e76\u6784\u5efa\u5173\u4e8e\u4e2a\u6027\u5316\u6cbb\u7597\u7b56\u7565\u6307\u5bfc\u7684\u6cbb\u7597\u53d8\u91cf\u6548\u5e94\u7684\u7f6e\u4fe1\u533a\u95f4\u3002\u8fd9\u79cdBootstrap\u7a0b\u5e8f\u53ef\u4ee5\u8003\u8651\u5230\u6570\u636e\u4e2d\u7684\u53f3\u4fa7\u622a\u5c3e\u751f\u5b58\u60c5\u51b5\uff0c\u4ece\u800c\u66f4\u51c6\u786e\u5730\u8fdb\u884c\u7edf\u8ba1\u63a8\u65ad\u3002<\/span><\/p>\n<\/li>\n

                  3. \n

                    \u9488\u5bf9\u52a8\u6001\u6cbb\u7597\u65b9\u6848\u7684\u8bc4\u4f30\u63d0\u51fa\u4e86\u4e00\u4e2a\u6846\u67b6\uff0c\u8be5\u6846\u67b6\u53ef\u4ee5\u6839\u636e\u60a3\u8005\u7684\u4e34\u5e8a\u53cd\u5e94\u548c\u5065\u5eb7\u7ed3\u679c\u8f68\u8ff9\uff0c\u9488\u5bf9\u6162\u6027\u75be\u75c5\u7b49\u6162\u6027\u75be\u75c5\u7684\u4e2a\u6027\u5316\u6cbb\u7597\u51b3\u7b56\u5236\u5b9a\u4e00\u7cfb\u5217\u6cbb\u7597\u51b3\u7b56\u89c4\u5219\u3002\u8fd9\u79cd\u52a8\u6001\u6cbb\u7597\u65b9\u6848\u53ef\u4ee5\u6839\u636e\u4e2a\u4f53\u7684\u533b\u7597\u5386\u53f2\u4e3a\u6bcf\u4e2a\u4eba\u91cf\u8eab\u5b9a\u5236\u6cbb\u7597\uff0c\u4ee5\u6700\u5927\u5316\u671f\u671b\u7684\u5065\u5eb7\u7ed3\u679c\u3002<\/span><\/p>\n<\/li>\n

                  4. \n
                    \u89e3\u51b3\u4e86\u5728\u4e2a\u6027\u5316\u6cbb\u7597\u51b3\u7b56\u4e2d\u7684\u4e00\u4e2a\u6311\u6218\uff0c\u5373\u5982\u4f55\u5e73\u8861\u591a\u4e2a\u7ade\u4e89\u6027\u7684\u5065\u5eb7\u7ed3\u679c\u3002\u6587\u7ae0\u63d0\u51fa\u7684\u65b9\u6cd5\u53ef\u4ee5\u5728\u8003\u8651\u591a\u4e2a\u5065\u5eb7\u7ed3\u679c\u7684\u60c5\u51b5\u4e0b\u8fdb\u884c\u51b3\u7b56\uff0c\u5e76\u5c06\u4e0d\u540c\u7684\u5065\u5eb7\u7ed3\u679c\u4e4b\u95f4\u7684\u6743\u8861\u7eb3\u5165\u5230\u51b3\u7b56\u6a21\u578b\u4e2d\uff0c\u4f8b\u5982\u5728\u6cbb\u7597\u51b3\u7b56\u4e2d\u5e73\u8861\u7597\u6548\u548c\u6bd2\u526f\u4f5c\u7528\u3002\u8fd9\u79cd\u65b9\u6cd5\u80fd\u591f\u66f4\u597d\u5730\u5e94\u5bf9\u591a\u76ee\u6807\u51b3\u7b56\u7684\u9700\u6c42\uff0c\u63d0\u9ad8\u4e2a\u6027\u5316\u6cbb\u7597\u51b3\u7b56\u7684\u6548\u679c\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

                    <\/section>\n
                    \u6574\u7406\u8005\uff1a\u8d21\u7545<\/span><\/section>\n

                    <\/span><\/section>\n

                    <\/span><\/section>\n

                    \n
                    \n
                    \n
                    \n
                    \n

                    <\/section>\n

                    <\/section>\n
                    \n
                    <\/svg><\/section>\n<\/section>\n<\/section>\n
                    \n

                    \u6570\u636e\u968f\u673a\u7f3a\u5931\u4e0b\u7684\u4e09\u91cd\u9c81\u68d2\u503e\u5411\u6027\u5206\u6570\u4f30\u8ba1<\/strong><\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                    \n

                    <\/section>\n

                    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                    <\/p>\n

                    \n
                    \u4e3b\u9898: Triply Robust Propensity Score Estimation Under Missing at Random<\/span><\/section>\n
                    \u4e3b\u8bb2\u4eba\uff1aJae-Kwang Kim<\/span><\/section>\n
                    \u7231\u8377\u534e\u5dde\u7acb\u5927\u5b66\u7edf\u8ba1\u7cfb\u6559\u6388\uff0c\u662f\u7f8e\u56fd\u7edf\u8ba1\u534f\u4f1a\uff08ASA\uff09\u548c\u56fd\u9645\u6570\u7406\u7edf\u8ba1\u5b66\u4f1a\uff08IMS\uff09\u4f1a\u58eb\uff0c\u5e76\u88ab\u7231\u8377\u534e\u5dde\u7acb\u5927\u5b66\u6587\u7406\u5b66\u9662\u8bc4\u4e3a\u6587\u7406\u5b66\u9662\u9662\u957f\u6559\u6388\u3002\u76ee\u524d\uff0cKim\u535a\u58eb\u62c5\u4efb\u97e9\u56fd\u56fd\u9645\u7edf\u8ba1\u5b66\u4f1a\uff08KISS\uff092023-2024 \u5e74\u5ea6\u4e3b\u5e2d\u3002Kim\u535a\u58eb\u7684\u4e3b\u8981\u7814\u7a76\u9886\u57df\u662f\u8c03\u67e5\u62bd\u6837\u548c\u7f3a\u5931\u6570\u636e\u5206\u6790\uff0c\u6700\u8fd1\u4e5f\u5728\u7814\u7a76\u4e00\u4e9b\u534a\u53c2\u6570\u4f30\u8ba1\u548c\u56e0\u679c\u63a8\u65ad\u7684\u76f8\u5173\u8bfe\u9898\u3002<\/span><\/section>\n
                    \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/sites.google.com\/view\/jaekwangkim\/home<\/span><\/section>\n<\/blockquote>\n


                    <\/h2>\n

                    \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n

                    Wang, H., Kim, J. K., Han, J., & Lee, Y. (2023). Triply robust estimation under missing at random. arXiv preprint arXiv:2306.15173<\/em>.<\/span><\/p>\n

                    \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                    \u6458\u8981<\/span><\/strong><\/h3>\n

                    \u7edf\u8ba1\u5b66\u7684\u8bb8\u591a\u9886\u57df\u7ecf\u5e38\u4f1a\u9047\u5230\u7f3a\u5931\u6570\u636e\u7684\u60c5\u51b5\u3002\u503e\u5411\u6027\u5206\u6570\u52a0\u6743\u6cd5\u662f\u5904\u7406\u7f3a\u5931\u6570\u636e\u7684\u5e38\u7528\u65b9\u6cd5\u3002\u503e\u5411\u6027\u5206\u6570\u91c7\u7528\u7684\u662f\u54cd\u5e94\u503e\u5411\u6a21\u578b\uff0c\u4f46\u5f53\u7f3a\u5931\u6570\u636e\u65f6\uff0c\u6a21\u578b\u8bbe\u5b9a\u7684\u6b63\u786e\u6027\u53ef\u80fd\u9762\u4e34\u6311\u6218\u3002\u5f53\u7ed3\u679c\u56de\u5f52\u6a21\u578b\u6216\u503e\u5411\u6027\u5206\u6570\u6a21\u578b\u90fd\u8bbe\u5b9a\u6b63\u786e\u65f6\uff0c\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u80fd\u4fdd\u8bc1\u4f30\u8ba1\u7ed3\u679c\u7684\u4e00\u81f4\u6027\u3002\u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u9996\u5148\u5229\u7528\u4fe1\u606f\u6295\u5f71\u6765\u5f00\u53d1\u95f4\u63a5\u6a21\u578b\u6821\u51c6\u7ea6\u675f\u4e0b\u7684\u9ad8\u6548\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\u3002\u901a\u8fc7\u5728\u4f30\u8ba1\u56de\u5f52\u53c2\u6570\u65f6\u65bd\u52a0\u5185\u90e8\u504f\u5dee\u6821\u51c6\u6761\u4ef6\uff0c\u5f97\u5230\u7684\u503e\u5411\u6027\u5206\u6570\u4f30\u8ba1\u5668\u53ef\u4ee5\u7b49\u4ef7\u5730\u8868\u793a\u4e3a\u53cc\u91cd\u7a33\u5065\u56de\u5f52\u4f30\u8ba1\u5668\u3002\u6b64\u5916\uff0c\u6211\u4eec\u8fd8\u5bf9\u4fe1\u606f\u6295\u5f71\u8fdb\u884c\u4e86\u6cdb\u5316\uff0c\u4ee5\u5b9e\u73b0\u5bf9\u79bb\u7fa4\u70b9\u7684\u7a33\u5065\u4f30\u8ba1\uff0c\u5e76\u5c55\u793a\u4e86\u4e00\u4e9b\u6e10\u8fdb\u5206\u6790\u7684\u7ed3\u679c\u3002\u865a\u62df\u5b9e\u9a8c\u7814\u7a76\u8bc1\u5b9e\uff0c\u6211\u4eec\u63d0\u51fa\u7684\u65b9\u6cd5\u4e0d\u4ec5\u53ef\u4ee5\u9488\u5bf9\u8fdd\u53cd\u5404\u79cd\u6a21\u578b\u5047\u8bbe\u7684\u60c5\u51b5\u8fdb\u884c\u7a33\u5065\u63a8\u65ad\uff0c\u8fd8\u53ef\u4ee5\u9488\u5bf9\u79bb\u7fa4\u70b9\u8fdb\u884c\u7a33\u5065\u63a8\u65ad\u3002\u771f\u5b9e\u6570\u636e\u5e94\u7528\u6848\u4f8b\u9009\u53d6\u4e86\u6c34\u571f\u4fdd\u6301\u6548\u679c\u8bc4\u4f30\u9879\u76ee\u7684\u6570\u636e\u3002<\/span><\/p>\n

                    \u8d21\u732e<\/span><\/strong><\/h3>\n

                    \u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u4e3a\u6570\u636e\u968f\u673a\u7f3a\u5931\uff08missing at random\uff09\u60c5\u51b5\u4e0b\u7684\u53cc\u91cd\u7a33\u5065\u503e\u5411\u6027\u5f97\u5206\u4f30\u8ba1\u5efa\u7acb\u4e86\u4e00\u4e2a\u7edf\u4e00\u6846\u67b6\u3002\u5177\u4f53\u5730\uff0c\u6211\u4eec\u5c06\u4fe1\u606f\u6295\u5f71\u5e94\u7528\u4e8e\u534f\u53d8\u91cf\u5e73\u8861\u7ea6\u675f\u4e0b\u7684\u503e\u5411\u6027\u5206\u6570\u52a0\u6743\u95ee\u9898\u3002<\/span><\/p>\n

                    \u65bd\u52a0\u534f\u53d8\u91cf\u5e73\u8861\u7ea6\u675f\u53ef\u4ee5\u7406\u89e3\u4e3a\u95f4\u63a5\u6a21\u578b\u6821\u51c6\uff0c\u8fd9\u662f\u503e\u5411\u6027\u5206\u6570\u52a0\u6743\u4f30\u8ba1\u5668\u5b9e\u73b0\u53cc\u91cd\u7a33\u5065\u6027\u7684\u5173\u952e\u6761\u4ef6\u3002\u7531\u6b64\u5f97\u5230\u7684\u503e\u5411\u6027\u5206\u6570\u52a0\u6743\u4f30\u8ba1\u5668\u53ef\u4ee5\u88ab\u8868\u793a\u4e3a\u56de\u5f52\u4f30\u8ba1\u5668\uff0c\u4f7f\u7528\u5e73\u8861\u51fd\u6570\u4f5c\u4e3a\u7ed3\u679c\u56de\u5f52\u6a21\u578b\u4e2d\u7684\u534f\u53d8\u91cf\u3002\u8fd9\u79cd\u4ee3\u6570\u7b49\u4ef7\u5173\u7cfb\u88ab\u79f0\u4e3a “\u81ea\u6548\u7387”\uff08self-efficiency\uff09\u3002<\/span><\/p>\n

                    \u7531\u4e8e\u7ed3\u679c\u53d8\u91cf\u4e2d\u7684\u79bb\u7fa4\u70b9\u5f80\u5f80\u4f1a\u5927\u5927\u964d\u4f4e\u4f30\u8ba1\u7ed3\u679c\u7684\u6548\u7387\uff0c\u6211\u4eec\u5f00\u53d1\u4e86\u4e00\u79cd\u57fa\u4e8e<\/span><\/span>\u03b3<\/span>\u5e42\u6563\u5ea6\u7684\u79bb\u7fa4\u70b9\u7a33\u5065\u56de\u5f52\u4f30\u7b97\u65b9\u6cd5\uff0c\u7136\u540e\u901a\u8fc7\u81ea\u6548\u7387\u6761\u4ef6\u5c06\u5176\u7b49\u4ef7\u8868\u793a\u4e3a\u503e\u5411\u6027\u5206\u6570\u52a0\u6743\u4f30\u8ba1\u5668\u3002<\/span><\/section>\n

                    <\/span><\/section>\n
                    \u6574\u7406\u8005\uff1a\u590f\u4e1a\u4f1f<\/span><\/section>\n


                    <\/h1>\n


                    <\/span><\/p>\n

                    \n
                    \n
                    \n
                    \n
                    \n

                    <\/section>\n

                    <\/section>\n
                    \n
                    <\/svg><\/section>\n<\/section>\n<\/section>\n
                    \n

                    \u5b58\u5728\u672a\u89c2\u5bdf\u4e2d\u4ecb-\u7ed3\u679c\u6df7\u6742\u4e0b\u7684\u7a00\u758f\u4e2d\u4ecb\u5206\u6790<\/strong><\/p>\n<\/section>\n

                    \n

                    <\/section>\n

                    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                    <\/p>\n

                    \n
                    \u4e3b\u9898\uff1aSparse Mediation Analysis with Unmeasured Mediator-outcome Confounding<\/span><\/section>\n
                    \u4e3b\u8bb2\u4eba\uff1aWei Li<\/span><\/section>\n
                    \u4e2d\u56fd\u4eba\u6c11\u5927\u5b66\u7edf\u8ba1\u5b66\u9662\u526f\u6559\u6388\u3002\u76ee\u524d\u62c5\u4efb\u4e2d\u56fd\u73b0\u573a\u7edf\u8ba1\u7814\u7a76\u4f1a\u56e0\u679c\u63a8\u65ad\u5206\u4f1a\u526f\u79d8\u4e66\u957f\uff0c\u4ee5\u53ca\u5317\u4eac\u751f\u7269\u533b\u5b66\u7edf\u8ba1\u4e0e\u6570\u636e\u7ba1\u7406\u7814\u7a76\u4f1a\u7406\u4e8b\u3002\u7814\u7a76\u65b9\u5411\u4e3a\u56e0\u679c\u63a8\u65ad\u3001\u7f3a\u5931\u6570\u636e\u3001\u6d4b\u91cf\u8bef\u5dee\u3001\u9ad8\u7ef4\u6570\u636e\u3001\u751f\u7269\u7edf\u8ba1\u3002\u8054\u7cfb\u90ae\u7bb1\uff1aweilistat@ruc.edu.cn<\/span><\/section>\n
                    \u4e2a\u4eba\u4e3b\u9875\uff1ahttp:\/\/stat.ruc.edu.cn\/Home\/People\/Faculty\/f09678c441cd4803a8432f1521904807.htm<\/span><\/section>\n<\/blockquote>\n


                    <\/h2>\n

                    \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n
                    \u56e0\u679c\u4e2d\u4ecb\u5206\u6790\u65e8\u5728\u7814\u7a76\u88ab\u79f0\u4e3a\u4e2d\u4ecb\u56e0\u5b50\u7684\u4e2d\u95f4\u56e0\u7d20\u5982\u4f55\u8c03\u8282\u6cbb\u7597\u5bf9\u7ed3\u679c\u7684\u56e0\u679c\u6548\u5e94\u3002\u968f\u7740\u5404\u5b66\u79d1\u5bf9\u5927\u91cf\u6f5c\u5728\u4e2d\u4ecb\u56e0\u7d20\u7684\u6d4b\u91cf\u8d8a\u6765\u8d8a\u591a\uff0c\u4eba\u4eec\u63d0\u51fa\u4e86\u5bf9\u8bb8\u591a\u751a\u81f3\u9ad8\u7ef4\u7684\u4e2d\u4ecb\u56e0\u7d20\u8fdb\u884c\u4e2d\u4ecb\u5206\u6790\u7684\u65b9\u6cd5\u3002\u7136\u800c\uff0c\u8fd9\u4e9b\u65b9\u6cd5\u901a\u5e38\u5047\u5b9a\u4e2d\u4ecb\u56e0\u5b50\u4e0e\u7ed3\u679c\u4e4b\u95f4\u4e0d\u5b58\u5728\u672a\u6d4b\u91cf\u7684\u6df7\u6742\u56e0\u5b50\u3002\u672c\u6587\u5141\u8bb8\u5b58\u5728\u8fd9\u79cd\u6df7\u6742\u56e0\u5b50\uff0c\u5e76\u63d0\u4f9b\u4e86\u4e00\u79cd\u5728\u7ed3\u6784\u7b49\u5f0f\u6a21\u578b\u6846\u67b6\u4e0b\u540c\u65f6\u89e3\u51b3\u8bc6\u522b\u548c\u4e2d\u4ecb\u9009\u62e9\u95ee\u9898\u7684\u65b9\u6cd5\u3002\u8bc6\u522b\u7b56\u7565\u5305\u62ec\u6839\u636e\u591a\u4e2d\u4ecb\u56e0\u5b50\u7684\u6f5c\u5728\u56e0\u5b50\u6a21\u578b\uff0c\u4e3a\u672a\u6d4b\u91cf\u7684\u6df7\u6742\u56e0\u7d20\u6784\u5efa\u4e00\u4e2a\u4f2a\u4ee3\u7406\u53d8\u91cf\u3002\u5229\u7528\u8be5\u4ee3\u7406\u53d8\u91cf\uff0c\u63d0\u51fa\u4e86\u4e00\u79cd\u90e8\u5206\u60e9\u7f5a\u8fc7\u7a0b\uff0c\u4ee5\u9009\u62e9\u5bf9\u7ed3\u679c\u5177\u6709\u975e\u96f6\u56e0\u679c\u6548\u5e94\u7684\u4e2d\u4ecb\u56e0\u5b50\u3002\u7531\u6b64\u5f97\u51fa\u7684\u4f30\u8ba1\u7ed3\u679c\u662f\u4e00\u81f4\u7684\uff0c\u5e76\u4e14\u975e\u96f6\u53c2\u6570\u7684\u4f30\u8ba1\u7ed3\u679c\u662f\u6e10\u8fd1\u6b63\u6001\u7684\u3002\u865a\u62df\u5b9e\u9a8c\u7814\u7a76\u8868\u660e\uff0c\u4e0e\u5176\u4ed6\u73b0\u6709\u65b9\u6cd5\u76f8\u6bd4\uff0c\u6240\u63d0\u51fa\u7684\u65b9\u6cd5\u6027\u80fd\u66f4\u4e3a\u4f18\u8d8a\u3002\u6700\u540e\uff0c\u8fd9\u79cd\u65b9\u6cd5\u88ab\u5e94\u7528\u4e8e\u57fa\u56e0\u7ec4\u6570\u636e\uff0c\u53ef\u80fd\u4f5c\u4e3a\u5c0f\u9f20\u80a5\u80d6\u75c7\u9057\u4f20\u53d8\u5f02\u6548\u5e94\u7684\u4e2d\u4ecb\u57fa\u56e0\u8868\u8fbe\u662f\u53ef\u8bc6\u522b\u7684\u3002<\/span><\/section>\n

                    <\/section>\n
                    \u6574\u7406\u8005\uff1a\u590f\u4e1a\u4f1f<\/span><\/section>\n


                    <\/h1>\n


                    <\/span><\/p>\n

                    \n
                    \n
                    \n
                    \n
                    \n

                    <\/section>\n

                    <\/section>\n
                    \n
                    <\/svg><\/section>\n<\/section>\n<\/section>\n
                    \n

                    \u57fa\u4e8e\u8fd1\u7aef\u56e0\u679c\u5b66\u4e60\u5f02\u8d28\u6cbb\u7597\u6548\u5e94<\/strong><\/strong><\/strong><\/p>\n<\/section>\n

                    \n

                    <\/section>\n

                    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                    <\/p>\n

                    \n
                    \u4e3b\u9898\uff1aProximal Causal Learning of Heterogeneous Treatment Effects<\/span><\/section>\n
                    \u4e3b\u8bb2\u4eba\uff1aYifan Cui<\/span><\/section>\n
                    \u6d59\u6c5f\u5927\u5b66\u6570\u636e\u79d1\u5b66\u7cfb\u52a9\u7406\u6559\u6388\u3002\u5728\u4ed6\u52a0\u5165\u6d59\u6c5f\u5927\u5b66\u4e4b\u524d\uff0c\u66fe\u4efb\u65b0\u52a0\u5761\u56fd\u7acb\u5927\u5b66\u7edf\u8ba1\u4e0e\u6570\u636e\u79d1\u5b66\u7cfb\u52a9\u7406\u6559\u6388\u3002\u66fe\u5728\u5bbe\u5915\u6cd5\u5c3c\u4e9a\u5927\u5b66\u6c83\u987f\u5546\u5b66\u9662\u62c5\u4efb\u535a\u58eb\u540e\u7814\u7a76\u5458\uff0c\u4ee5\u53ca\u5728\u5317\u5361\u7f57\u6765\u7eb3\u5927\u5b66\u6559\u5802\u5c71\u5206\u6821\u83b7\u5f97\u535a\u58eb\u5b66\u4f4d\u3002\u5f53\u9009\u4e86IS(\u56fd\u9645\u7edf\u8ba1\u5b66\u4f1a) Elected Member\uff0c\u5165\u9009\u798f\u5e03\u65af\u4e9a\u6d32U30\u6770\u51fa\u9752\u5e74\uff0c\u73b0\u62c5\u4efbBiometrical Journal\u7684Associate Editor\u4ee5\u53caJournal of Machine LearningBesearch\u7684editorial board reviewer.<\/span><\/section>\n
                    \u4e2a\u4eba\u4e3b\u9875\uff1ahttps:\/\/sites.google.com\/view\/yifancui\/menu\/home<\/span><\/section>\n<\/blockquote>\n


                    <\/h2>\n

                    \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                    Sverdrup, E., & Cui, Y. (2023). Proximal Causal Learning of Heterogeneous Treatment Effects. arXiv preprint arXiv:2301.10913<\/em>.<\/span><\/section>\n


                    <\/h2>\n

                    \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                    \u6458\u8981<\/span><\/strong><\/h3>\n

                    \u9ad8\u6548\u7075\u6d3b\u5730\u4f30\u8ba1\u5f02\u8d28\u6cbb\u7597\u6548\u5e94\u662f\u4e00\u9879\u91cd\u8981\u4efb\u52a1\uff0c\u6d89\u53ca\u4ece\u533b\u5b66\u5230\u5e02\u573a\u8425\u9500\u7b49\u591a\u79cd\u9886\u57df\u3002\u76ee\u524d\u5df2\u7ecf\u6709\u8bb8\u591a\u6761\u4ef6\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u4f30\u8ba1\u5668\u88ab\u63d0\u51fa\uff0c\u7136\u800c\u8fd9\u4e9b\u65b9\u6cd5\u901a\u5e38\u4f9d\u8d56\u4e8e\u89c2\u6d4b\u534f\u53d8\u91cf\u8db3\u4ee5\u8bc1\u5b9e\u6761\u4ef6\u53ef\u4ea4\u6362\u6027\u7684\u524d\u63d0\u5047\u8bbe\u3002\u6211\u4eec\u53d7 R-learner \u548c DR-learner \u7684\u542f\u53d1\u63d0\u51fa\u4e86 P-learner\uff0c\u8fd9\u662f\u4e00\u79cd\u5b9a\u5236\u7684\u4e24\u9636\u6bb5\u635f\u5931\u51fd\u6570\uff0c\u7528\u4e8e\u5728\u4ee5\u89c2\u6d4b\u534f\u53d8\u91cf\u4e3a\u6761\u4ef6\u65f6\u53ef\u4ea4\u6362\u6027\u4e0d\u53ef\u4fe1\u7684\u8bbe\u5b9a\u4e0b\u5b66\u4e60\u5f02\u8d28\u6cbb\u7597\u6548\u5e94\u3002\u6211\u4eec\u5e0c\u671b\u4f9d\u9760\u4e00\u4e9b\u66ff\u4ee3\u53d8\u91cf\u8fdb\u884c\u56e0\u679c\u63a8\u65ad\u3002\u6211\u4eec\u63d0\u51fa\u7684\u4f30\u8ba1\u5668\u53ef\u4ee5\u901a\u8fc7\u73b0\u6210\u7684\u6700\u5c0f\u5316\u635f\u5931\u7684\u673a\u5668\u5b66\u4e60\u65b9\u6cd5\u5b9e\u73b0\u3002\u5728\u6838\u56de\u5f52\u7684\u60c5\u51b5\u4e0b\uff0c\u53ea\u8981\u6270\u52a8\u6210\u5206\u5f97\u5230\u5408\u7406\u7684\u4f30\u8ba1\uff0c\u4f30\u8ba1\u8bef\u5dee\u5c31\u80fd\u53d6\u5230\u6700\u4f18\u7684\u754c\u3002<\/span><\/p>\n

                    \u8d21\u732e<\/span><\/strong><\/h3>\n
                    \u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u6761\u4ef6\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u4f30\u8ba1\u5668\uff0c\u5b83\u5141\u8bb8\u6211\u4eec\u5728\u6761\u4ef6\u53ef\u4ea4\u6362\u6027\u5931\u6548\uff0c\u4f46\u5df2\u7ecf\u6d4b\u91cf\u4e86\u4e00\u7ec4\u5145\u5206\u66ff\u4ee3\u53d8\u91cf\u7684\u60c5\u51b5\u4e0b\u4f30\u8ba1\u56e0\u679c\u6548\u5e94\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u53d7\u901a\u7528 Neyman-\u6b63\u4ea4\u635f\u5931\u51fd\u6570\u7684\u542f\u53d1\uff0c\u8be5\u51fd\u6570\u5c06\u6270\u52a8\u4f30\u8ba1\u548c\u6761\u4ef6\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u4f30\u8ba1\u89e3\u8026\u4e3a\u4e24\u4e2a\u9636\u6bb5\uff0c\u53ef\u901a\u8fc7\u7075\u6d3b\u7684\u57fa\u4e8e\u635f\u5931\u6700\u5c0f\u5316\u7684\u673a\u5668\u5b66\u4e60\u5de5\u5177\u8fdb\u884c\u4f30\u8ba1\uff0c\u5176\u4e2d\u540e\u4e00\u9636\u6bb5\u5bf9\u7b2c\u4e00\u9636\u6bb5\u4ea7\u751f\u7684\u4f30\u8ba1\u8bef\u5dee\u7684\u654f\u611f\u5ea6\u8f83\u4f4e\u3002\u6211\u4eec\u7684\u8d21\u732e\u5728\u4e8e\u5c06\u8fd9\u79cd\u7075\u6d3b\u7684\u6761\u4ef6\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u4f30\u8ba1\u7b56\u7565\u6269\u5c55\u5230\u8fd1\u7aef\u56e0\u679c\u63a8\u65ad\u6846\u67b6\uff0c\u6240\u63d0\u51fa\u7684\u635f\u5931\u51fd\u6570\u4f9d\u8d56\u4e8e\u53cc\u91cd\u7a33\u5065\u5206\u6570\u3002\u8fd9\u4e9b\u65b9\u6cd5\u4e5f\u53ef\u7528\u4e8e\u5bf9\u6761\u4ef6\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u7684\u4f4e\u7ef4\u8868\u793a\u8fdb\u884c\u534a\u53c2\u6570\u9ad8\u6548\u4f30\u8ba1\u548c\u63a8\u65ad\uff0c\u5982\u6700\u4f18\u7ebf\u6027\u9884\u6d4b\u6216\u52a0\u6743\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u3002<\/span><\/section>\n

                    <\/section>\n
                    \u6574\u7406\u8005\uff1a\u590f\u4e1a\u4f1f<\/span><\/section>\n


                    <\/h1>\n

                    <\/span><\/section>\n

                    \n
                    \n
                    \n
                    \n
                    \n

                    <\/section>\n

                    <\/section>\n
                    \n
                    <\/svg><\/section>\n<\/section>\n<\/section>\n
                    \n

                    \u57fa\u4e8e\u9ad8\u9636\u7d2f\u79ef\u91cf\u7684\u5e26\u6709\u6f5c\u53d8\u91cf\u7684\u56e0\u679c\u53d1\u73b0<\/span><\/strong><\/p>\n<\/section>\n

                    \n

                    <\/section>\n

                    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                    <\/p>\n

                    \n
                    \u4e3b\u8bb2\u4eba\uff1aFan Yang<\/span><\/section>\n
                    \u5e7f\u4e1c\u5de5\u4e1a\u5927\u5b66\u8bb2\u5e08\u3002\u5206\u522b\u4e8e2015\u5e74\u548c2020\u5e74\u83b7\u5f97\u5e7f\u4e1c\u5de5\u4e1a\u5927\u5b66\u8ba1\u7b97\u673a\u79d1\u5b66\u5b66\u58eb\u5b66\u4f4d\u548c\u8ba1\u7b97\u673a\u5e94\u7528\u5de5\u7a0b\u535a\u58eb\u5b66\u4f4d\u30022018-2019\u5e74\uff0c\u7f8e\u56fd\u5bbe\u5915\u6cd5\u5c3c\u4e9a\u5dde\u5339\u5179\u5821\u5361\u5185\u57fa\u6885\u9686\u5927\u5b66\u7684\u8bbf\u95ee\u5b66\u751f\u3002\u7814\u7a76\u5174\u8da3\u5305\u62ec\u56e0\u679c\u53d1\u73b0\u53ca\u5176\u5e94\u7528\u3002<\/span><\/section>\n
                    \u4e2a\u4eba\u4e3b\u9875:<\/span><\/section>\n
                    htpss\/\/yzw.gdut.edu,cilinfo\/1120\/6724.htm<\/span><\/section>\n<\/blockquote>\n


                    <\/h2>\n

                    \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                    Cai, R., Huang, Z., Chen, W., Hao, Z., & Zhang, K. (2023). Causal Discovery with Latent Confounders Based on Higher-Order Cumulants. arXiv preprint arXiv:2305.19582<\/em>.<\/span><\/section>\n


                    <\/h2>\n

                    \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                    \u6458\u8981<\/span><\/strong><\/h3>\n

                    \u5177\u6709\u6f5c\u5728\u6df7\u6dc6\u56e0\u7d20\u7684\u56e0\u679c\u53d1\u73b0\u662f\u8bb8\u591a\u79d1\u5b66\u9886\u57df\u4e2d\u91cd\u8981\u4f46\u5177\u6709\u6311\u6218\u6027\u7684\u4efb\u52a1\u3002\u5c3d\u7ba1\u5728\u67d0\u4e9b\u9886\u57df\u4e2d\u57fa\u4e8e\u8fc7\u5ea6\u5b8c\u5907\u72ec\u7acb\u6210\u5206\u5206\u6790\uff08OICA\uff09\u7684\u65b9\u6cd5\u53d6\u5f97\u4e86\u6210\u529f\uff0c\u4f46\u5b83\u4eec\u5728\u8ba1\u7b97\u4e0a\u5f88\u6602\u8d35\uff0c\u5e76\u4e14\u5bb9\u6613\u9677\u5165\u5c40\u90e8\u6700\u4f18\u89e3\u3002\u6211\u4eec\u6ce8\u610f\u5230\u6709\u8da3\u7684\u662f\uff0c\u901a\u8fc7\u5229\u7528\u9ad8\u9636\u7d2f\u79ef\u91cf\uff0c\u5b58\u5728\u4e00\u79cd\u7279\u5b9a\u60c5\u51b5\u4e0b\uff08\u4f8b\u5982\uff0c\u5f53\u6df7\u5408\u8fc7\u7a0b\u9075\u5faa\u4e00\u79cd\u6f5c\u5728\u6210\u5206\u7ed3\u6784\u65f6\uff09\uff0cOICA\u7684\u5c01\u95ed\u5f62\u5f0f\u89e3\u51b3\u65b9\u6848\u3002\u9274\u4e8e\u5c01\u95ed\u5f62\u5f0f\u89e3\u51b3\u65b9\u6848\u5bf9\u5e94\u4e8eOne-Latent-Component\u7ed3\u6784\uff0c\u6211\u4eec\u5236\u5b9a\u4e86\u4e00\u79cd\u65b9\u6cd5\u6765\u4f7f\u7528\u9ad8\u9636\u7d2f\u79ef\u91cf\u4f30\u8ba1\u6df7\u5408\u77e9\u9635\uff0c\u5e76\u8fdb\u4e00\u6b65\u63d0\u51fa\u4e86\u53ef\u6d4b\u8bd5\u7684One-Latent-Component\u6761\u4ef6\u6765\u8bc6\u522b\u6f5c\u5728\u53d8\u91cf\u5e76\u786e\u5b9a\u56e0\u679c\u987a\u5e8f\u3002\u901a\u8fc7\u8fed\u4ee3\u5730\u79fb\u9664\u5171\u540c\u8bc6\u522b\u7684\u6f5c\u5728\u6210\u5206\uff0c\u6211\u4eec\u6210\u529f\u5730\u5c06\u7ed3\u679c\u6269\u5c55\u5230Multi-Latent-Component\u7ed3\u6784\uff0c\u5e76\u6700\u7ec8\u63d0\u4f9b\u4e86\u4e00\u79cd\u5b66\u4e60\u5177\u6709\u6f5c\u5728\u53d8\u91cf\u7684\u56e0\u679c\u7ed3\u6784\u7684\u5b9e\u7528\u4e14\u6e10\u8fd1\u6b63\u786e\u7684\u7b97\u6cd5\u3002\u5b9e\u9a8c\u7ed3\u679c\u8bf4\u660e\u4e86\u6240\u63d0\u65b9\u6cd5\u7684\u6e10\u8fd1\u6b63\u786e\u6027\u548c\u6709\u6548\u6027\u3002<\/span><\/p>\n

                    \u521b\u65b0<\/span><\/strong><\/h3>\n
                      \n
                    1. \n

                      \u89e3\u51b3\u4e86\u5177\u6709\u6f5c\u5728\u6df7\u6dc6\u56e0\u7d20\u7684\u56e0\u679c\u53d1\u73b0\u95ee\u9898\uff0c\u8fd9\u5728\u5b66\u4e60\u56e0\u679c\u7ed3\u6784\u4e2d\u901a\u5e38\u4f1a\u5f15\u53d1\u4e25\u91cd\u7684\u53ef\u8bc6\u522b\u6027\u95ee\u9898\u3002\u4f20\u7edf\u65b9\u6cd5\uff08\u5982FCI\u3001RFCI\u7b49\uff09\u5728\u89e3\u51b3\u8fd9\u4e00\u95ee\u9898\u65f6\u901a\u5e38\u53d7\u5230\u9a6c\u5c14\u79d1\u592b\u7b49\u4ef7\u7c7b\u7684\u9650\u5236\uff0c\u4f46\u672c\u6587\u5f15\u5165\u4e86\u7ebf\u6027\u548c\u975e\u9ad8\u65af\u6a21\u578b\uff0c\u4f7f\u5f97\u5373\u4f7f\u5b58\u5728\u6f5c\u5728\u6df7\u6dc6\u56e0\u7d20\uff0c\u4e5f\u80fd\u591f\u8bc6\u522b\u56e0\u679c\u7ed3\u6784\u3002<\/span><\/p>\n<\/li>\n

                    2. \n

                      \u5f15\u5165\u4e86\u4e00\u79cd\u79f0\u4e3aOne-Latent-Component\u7ed3\u6784\u7684\u7279\u5b9a\u60c5\u51b5\uff0c\u5176\u4e2d\u4e24\u4e2a\u89c2\u6d4b\u53d8\u91cf\u53d7\u5230\u4e00\u4e2a\u5171\u4eab\u7684\u6f5c\u5728\u6210\u5206\u7684\u76f4\u63a5\u5f71\u54cd\u3002\u901a\u8fc7\u4f7f\u7528\u9ad8\u9636\u7d2f\u79ef\u91cf\uff0c\u6587\u7ae0\u63d0\u4f9b\u4e86\u4e00\u79cdOICA\u7684\u5c01\u95ed\u5f62\u5f0f\u89e3\u51b3\u65b9\u6848\uff0c\u53ef\u4ee5\u7528\u4e8e\u4f30\u8ba1\u6f5c\u5728\u6df7\u6dc6\u56e0\u7d20\u7684\u56e0\u679c\u7cfb\u6570\u6bd4\u3002<\/span><\/p>\n<\/li>\n

                    3. \n

                      \u63d0\u51fa\u4e86\u4e00\u79cd\u65b9\u6cd5\uff0c\u53ef\u4ee5\u5728\u5df2\u77e5\u56e0\u679c\u7ed3\u6784\u7684\u60c5\u51b5\u4e0b\u4f30\u8ba1\u6df7\u5408\u77e9\u9635\uff0c\u5e76\u8fdb\u4e00\u6b65\u6062\u590d\u56e0\u679c\u5173\u7cfb\u3002\u8fd9\u4e00\u65b9\u6cd5\u901a\u8fc7\u63a2\u7d22\u6df7\u5408\u77e9\u9635\u4e2d\u5916\u751f\u548c\u5185\u751f\u53d8\u91cf\u4e4b\u95f4\u7684\u5173\u7cfb\uff0c\u5e76\u5229\u7528One-Latent-Component\u6761\u4ef6\u6765\u68c0\u9a8c\u5047\u8bbe\u7684\u56e0\u679c\u7ed3\u6784\u662f\u5426\u6b63\u786e\u3002<\/span><\/p>\n<\/li>\n

                    4. \n

                      \u6587\u7ae0\u8fd8\u63d0\u51fa\u4e86\u4e00\u79cd\u5c06\u4e0a\u8ff0\u65b9\u6cd5\u63a8\u5e7f\u5230Multi-LatentComponent\u7ed3\u6784\u7684\u601d\u8def\uff0c\u901a\u8fc7\u53bb\u9664\u5171\u4eab\u7684\u72ec\u7acb\u6210\u5206\uff0c\u8fdb\u4e00\u6b65\u6062\u590d\u5b8c\u6574\u7684\u56e0\u679c\u7ed3\u6784\u3002<\/span><\/p>\n<\/li>\n<\/ol>\n

                      \u603b\u7684\u6765\u8bf4\uff0c\u8fd9\u7bc7\u6587\u7ae0\u7684\u521b\u65b0\u70b9\u5728\u4e8e\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u65b9\u6cd5\uff0c\u53ef\u4ee5\u5728\u5b58\u5728\u6f5c\u5728\u6df7\u6dc6\u56e0\u7d20\u7684\u60c5\u51b5\u4e0b\u8fdb\u884c\u56e0\u679c\u53d1\u73b0\uff0c\u7279\u522b\u662f\u5728\u7ebf\u6027\u548c\u975e\u9ad8\u65af\u6a21\u578b\u7684\u80cc\u666f\u4e0b\u3002\u8fd9\u4e00\u65b9\u6cd5\u4e3a\u89e3\u51b3\u56e0\u679c\u7ed3\u6784\u5b66\u4e60\u4e2d\u7684\u53ef\u8bc6\u522b\u6027\u95ee\u9898\u63d0\u4f9b\u4e86\u65b0\u7684\u601d\u8def\uff0c\u5e76\u5728\u7279\u5b9a\u60c5\u51b5\u4e0b\u63d0\u4f9b\u4e86\u5c01\u95ed\u5f62\u5f0f\u7684\u89e3\u51b3\u65b9\u6848\u3002\u540c\u65f6\uff0c\u6587\u7ae0\u8fd8\u63a2\u8ba8\u4e86\u5982\u4f55\u5728\u5df2\u77e5\u56e0\u679c\u7ed3\u6784\u7684\u60c5\u51b5\u4e0b\u4f30\u8ba1\u6df7\u5408\u77e9\u9635\uff0c\u5e76\u63d0\u51fa\u4e86\u4e00\u79cd\u53ef\u884c\u7684\u65b9\u6cd5\u3002<\/span><\/section>\n

                      <\/section>\n
                      \u6574\u7406\u8005\uff1a\u82cf\u5a49\u5a77<\/span><\/section>\n


                      <\/h1>\n


                      <\/span><\/p>\n

                      \n
                      \n
                      \n
                      \n
                      \n

                      <\/section>\n

                      <\/section>\n
                      \n
                      <\/svg><\/section>\n<\/section>\n<\/section>\n
                      \n

                      \u5728\u4f7f\u7528\u5916\u90e8\u89c2\u5bdf\u6570\u636e\u7684\u4e34\u5e8a\u8bd5\u9a8c\u4e2d<\/strong><\/span><\/p>\n

                      \u5bf9\u957f\u671f\u7ed3\u679c\u7684\u6cbb\u7597\u6548\u679c\u8fdb\u884c\u8bc6\u522b\u548c\u4f30\u8ba1<\/strong><\/span><\/strong><\/strong><\/p>\n<\/section>\n

                      \n

                      <\/section>\n

                      <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


                      <\/p>\n

                      \n
                      \u4e3b\u9898\uff1aIdentification and Estimation of Treatment Effects on Long-term Outcomes in Clinical Trials with External Observational Data<\/span><\/section>\n
                      \u4e3b\u8bb2\u4eba\uff1aWenjie Hu \uff0c\u5317\u4eac\u5927\u5b66\u535a\u58eb\u751f<\/span><\/section>\n
                      Google\u5b66\u672f:<\/span><\/section>\n
                      https:\/\/scholar.google.com.hk\/citations?user=2FC7k-QAAAAJ&hl=en&oi=sra<\/span><\/section>\n<\/blockquote>\n


                      <\/h2>\n

                      \u5f15\u7528\u6587\u732e<\/span><\/strong><\/span><\/h2>\n
                      Hu, Wenjie, Xiaohua Zhou, and Peng Wu. “Identification and estimation of treatment effects on long-term outcomes in clinical trials with external observational data.” arXiv preprint arXiv:2208.10163<\/em> (2022).<\/span><\/section>\n


                      <\/h2>\n

                      \u5185\u5bb9<\/span><\/strong><\/span><\/h2>\n

                      \u6458\u8981<\/span><\/strong><\/h3>\n

                      \u5728\u751f\u7269\u533b\u5b66\u7814\u7a76\u4e2d\uff0c\u4f30\u8ba1\u836f\u7269\u5bf9\u6162\u6027\u75be\u75c5\u7684\u6548\u679c\u9700\u8981\u957f\u65f6\u95f4\u7684\u968f\u8bbf\u671f\uff0c\u800c\u8fd9\u5728\u968f\u673a\u4e34\u5e8a\u8bd5\u9a8c\uff08RCTs\uff09\u4e2d\u5f80\u5f80\u96be\u4ee5\u5b9e\u73b0\u3002\u7528\u77ed\u671f\u4ee3\u7528\u6307\u6807\u66ff\u4ee3\u957f\u671f\u7ed3\u679c\u6765\u8bc4\u4f30\u836f\u7269\u6548\u679c\u4f9d\u8d56\u4e8e\u4e25\u683c\u7684\u5047\u8bbe\uff0c\u800c\u5b9e\u8bc1\u7814\u7a76\u5f80\u5f80\u65e0\u6cd5\u6ee1\u8db3\u8fd9\u4e9b\u5047\u8bbe\u3002\u5728\u4e00\u4e2a\u80be\u810f\u75be\u75c5\u7814\u7a76\u7684\u52a8\u673a\u4e0b\uff0c\u672c\u6587\u7ed3\u5408\u4e86\u4e00\u4e2a\u6ca1\u6709\u89c2\u5bdf\u5230\u957f\u671f\u7ed3\u679c\u7684 RCT \u548c\u4e00\u4e2a\u89c2\u5bdf\u5230\u4f46\u53ef\u80fd\u5b58\u5728\u672a\u6d4b\u6df7\u6742\u56e0\u7d20\u7684\u89c2\u5bdf\u7814\u7a76\uff0c\u7814\u7a76\u4e86\u836f\u7269\u5bf9\u957f\u671f\u7ed3\u679c\u7684\u6548\u5e94\u3002\u5728\u4e00\u4e2a\u5f31\u4e8e\u5148\u524d\u6587\u732e\u7684\u5e73\u5747\u4ea4\u6362\u6027\u5047\u8bbe\u4e0b\uff0c\u672c\u6587\u786e\u5b9a\u4e86 RCT \u4e2d\u7684\u5e73\u5747\u6cbb\u7597\u6548\u5e94\uff0c\u5e76\u63a8\u5bfc\u4e86\u76f8\u5e94\u7684\u6709\u6548\u5f71\u54cd\u51fd\u6570\u548c\u534a\u53c2\u6570\u6548\u7387\u754c\u9650\u3002\u6b64\u5916\uff0c\u672c\u6587\u63d0\u51fa\u4e86\u4e00\u79cd\u5c40\u90e8\u9ad8\u6548\u7684\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\u548c\u5012\u6570\u6982\u7387\u52a0\u6743\uff08IPW\uff09\u4f30\u8ba1\u5668\u3002\u5982\u679c\u6240\u6709\u7684\u5de5\u4f5c\u6a21\u578b\u90fd\u88ab\u6b63\u786e\u6307\u5b9a\uff0c\u524d\u8005\u53ef\u4ee5\u8fbe\u5230\u534a\u53c2\u6570\u6548\u7387\u754c\u9650\uff0c\u4f46\u7531\u4e8e\u4ea4\u7ec7\u7684\u5de5\u4f5c\u6a21\u578b\uff0c\u8fd9\u53ef\u80fd\u5f88\u96be\u5b9e\u73b0\u3002\u540e\u8005\u5177\u6709\u66f4\u7b80\u5355\u7684\u5f62\u5f0f\uff0c\u5e76\u4e14\u9700\u8981\u66f4\u5c11\u7684\u6a21\u578b\u89c4\u8303\u3002\u4f7f\u7528\u4f30\u8ba1\u7684\u503e\u5411\u5206\u6570\u7684IPW\u4f30\u8ba1\u5668\u6bd4\u4f7f\u7528\u771f\u5b9e\u503e\u5411\u5206\u6570\u66f4\u9ad8\u6548\uff0c\u5e76\u5728\u79bb\u6563\u534f\u53d8\u91cf\u548c\u6709\u9650\u652f\u6301\u7684\u4ee3\u7528\u6307\u6807\u60c5\u51b5\u4e0b\u8fbe\u5230\u534a\u53c2\u6570\u6548\u7387\u754c\u9650\u3002\u8fd9\u4e24\u79cd\u4f30\u8ba1\u5668\u5747\u8868\u73b0\u51fa\u4e00\u81f4\u6027\u548c\u6e10\u8fd1\u6b63\u6001\u5206\u5e03\u3002\u672c\u6587\u8fdb\u884c\u4e86\u5927\u91cf\u7684\u6a21\u62df\u5b9e\u9a8c\u6765\u8bc4\u4f30\u6240\u63d0\u4f30\u8ba1\u5668\u7684\u6709\u9650\u6837\u672c\u6027\u80fd\u3002\u672c\u6587\u5e94\u7528\u6240\u63d0\u65b9\u6cd5\u6765\u4f30\u8ba1\u771f\u5b9e\u4e16\u754c\u6570\u636e\u5206\u6790\u4e2d\u53e3\u670d\u7f9f\u6c2f\u55b9\u5bf9\u80be\u529f\u80fd\u8870\u7aed\u7684\u7597\u6548\u3002<\/span><\/p>\n

                      \u521b\u65b0<\/span><\/strong><\/h3>\n
                        \n
                      1. \n

                        \u5f31\u5316\u5047\u8bbe\u6761\u4ef6\uff1a\u76f8\u6bd4\u73b0\u6709\u65b9\u6cd5\uff0c\u672c\u6587\u5728\u5e73\u5747\u53ef\u4ea4\u6362\u6027\u5047\u8bbe\u7684\u524d\u63d0\u4e0b\uff0c\u7814\u7a76\u4e86\u5728RCT\uff08\u968f\u673a\u5bf9\u7167\u8bd5\u9a8c\uff09\u4e2d\u65e0\u6cd5\u89c2\u5bdf\u5230\u957f\u671f\u7ed3\u679c\u7684\u60c5\u51b5\u4e0b\uff0c\u901a\u8fc7\u7ed3\u5408\u89c2\u5bdf\u7814\u7a76\u4e2d\u53ef\u4ee5\u89c2\u5bdf\u5230\u957f\u671f\u7ed3\u679c\u4f46\u53ef\u80fd\u5b58\u5728\u672a\u6d4b\u6df7\u6742\u53d8\u91cf\u7684\u6570\u636e\uff0c\u6765\u8bc6\u522b\u6cbb\u7597\u5bf9\u957f\u671f\u7ed3\u679c\u7684\u5f71\u54cd\u3002\u8fd9\u4e00\u5047\u8bbe\u6761\u4ef6\u8f83\u73b0\u6709\u65b9\u6cd5\u66f4\u5f31\u3002<\/span><\/p>\n<\/li>\n

                      2. \n

                        \u534a\u53c2\u6570\u6548\u7387\u8fb9\u754c\uff1a\u6587\u7ae0\u63a8\u5bfc\u51fa\u4e86\u76ee\u6807\u53c2\u6570\u7684\u534a\u53c2\u6570\u6548\u7387\u5f71\u54cd\u51fd\u6570\u548c\u534a\u53c2\u6570\u6548\u7387\u8fb9\u754c\u3002\u8fd9\u4e3a\u5728\u4f30\u8ba1RCT\u4e2d\u7684\u5e73\u5747\u6cbb\u7597\u6548\u5e94\u65f6\u63d0\u4f9b\u4e86\u7406\u8bba\u4e0a\u7684\u6548\u7387\u754c\u9650\u3002<\/span><\/p>\n<\/li>\n

                      3. \n
                        \u63d0\u51fa\u4e86\u4e24\u79cd\u4f30\u8ba1\u65b9\u6cd5\uff1a\u6587\u7ae0\u63d0\u51fa\u4e86\u4e00\u79cd\u5c40\u90e8\u9ad8\u6548\u7684\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\u548c\u4e00\u79cd\u5012\u6570\u6982\u7387\u52a0\u6743\uff08IPW\uff09\u4f30\u8ba1\u5668\u3002\u5c40\u90e8\u9ad8\u6548\u7684\u53cc\u91cd\u7a33\u5065\u4f30\u8ba1\u5668\u5728\u6240\u6709\u5de5\u4f5c\u6a21\u578b\u90fd\u6b63\u786e\u6307\u5b9a\u7684\u60c5\u51b5\u4e0b\uff0c\u8fbe\u5230\u4e86\u534a\u53c2\u6570\u6548\u7387\u8fb9\u754c\u3002\u5012\u6570\u6982\u7387\u52a0\u6743\u4f30\u8ba1\u5668\u5177\u6709\u66f4\u7b80\u5355\u7684\u5f62\u5f0f\uff0c\u5e76\u4e14\u9700\u8981\u66f4\u5c11\u7684\u6a21\u578b\u89c4\u8303\u3002\u4f7f\u7528\u4f30\u8ba1\u7684\u503e\u5411\u5f97\u5206\u7684IPW\u4f30\u8ba1\u5668\u6bd4\u4f7f\u7528\u771f\u5b9e\u503e\u5411\u5f97\u5206\u66f4\u9ad8\u6548\uff0c\u5e76\u5728\u79bb\u6563\u534f\u53d8\u91cf\u548c\u5177\u6709\u6709\u9650\u652f\u6301\u7684\u4ee3\u7406\u53d8\u91cf\u60c5\u51b5\u4e0b\u5b9e\u73b0\u4e86\u534a\u53c2\u6570\u6548\u7387\u8fb9\u754c\u3002\u8fd9\u4e24\u79cd\u4f30\u8ba1\u65b9\u6cd5\u5747\u8868\u73b0\u51fa\u4e00\u81f4\u6027\u548c\u6e10\u8fd1\u6b63\u6001\u5206\u5e03\u7684\u6027\u8d28\u3002<\/span><\/section>\n<\/li>\n<\/ol>\n

                        <\/section>\n
                        \u6574\u7406\u8005: \u8d21\u7545<\/span><\/section>\n


                        <\/p>\n


                        <\/p>\n

                        \n
                        \n
                        \n
                        \n

                        \u52a0\u5165\u56e0\u679c\u79d1\u5b66\u793e\u533a<\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n


                        <\/p>\n

                        \u7531\u667a\u6e90\u793e\u533a\u3001\u96c6\u667a\u4ff1\u4e50\u90e8\u8054\u5408\u4e3e\u529e\u7684<\/span>\u56e0\u679c\u79d1\u5b66\u4e0eCausal AI\u8bfb\u4e66\u4f1a\u7b2c\u4e09\u5b63<\/a>\uff0c\u4e3b\u8981\u9762\u5411\u4e24\u7c7b\u4eba\u7fa4\uff1a\u5982\u679c\u4f60\u4ece\u4e8b\u8ba1\u7b97\u673a\u76f8\u5173\u65b9\u5411\u7814\u7a76\uff0c\u5e0c\u671b\u4e3a\u4e0d\u540c\u9886\u57df\u5f15\u5165\u65b0\u7684\u8ba1\u7b97\u65b9\u6cd5\uff0c\u901a\u8fc7\u5927\u6570\u636e\u3001\u65b0\u7b97\u6cd5\u5f97\u5230\u65b0\u6210\u679c\uff0c\u53ef\u4ee5\u901a\u8fc7\u8bfb\u4e66\u4f1a\u5404\u4e2a\u9886\u57df\u7684\u6838\u5fc3\u56e0\u679c\u95ee\u9898\u4ecb\u7ecd\u548c\u8bba\u6587\u63a8\u8350\u5feb\u901f\u5165\u624b\uff1b\u5982\u679c\u4f60\u4ece\u4e8b\u5176\u4ed6\u7406\u5de5\u79d1\u6216\u4eba\u6587\u793e\u79d1\u9886\u57df\u7814\u7a76\uff0c\u4e5f\u53ef\u4ee5\u901a\u8fc7\u6240\u5c5e\u9886\u57df\u7684\u56e0\u679c\u7814\u7a76\u7efc\u8ff0\u4ecb\u7ecd\u548c\u7814\u8ba8\u5df2\u6709\u5de5\u4f5c\u7684\u793a\u4f8b\u4ee3\u7801\uff0c\u5728\u81ea\u5df1\u7684\u7814\u7a76\u4e2d\u5feb\u901f\u5f00\u59cb\u5c1d\u8bd5\u90e8\u7f72\u7ed3\u5408\u56e0\u679c\u7684\u7b97\u6cd5\u3002\u8bfb\u4e66\u4f1a\u76f4\u64ad\u5df2\u7ed3\u675f\uff0c\u6b22\u8fce\u52a0\u5165\u56e0\u679c\u79d1\u5b66\u793e\u533a\uff0c\u56de\u770b\u7b2c\u4e09\u5b63\u76f4\u64ad\uff0c\u52a0\u5165\u8ba8\u8bba\u3002<\/span><\/section>\n


                        <\/span><\/p>\n

                        <\/span><\/a><\/span><\/p>\n


                        <\/p>\n

                        \u8be6\u60c5\u8bf7\u89c1\uff1a<\/span><\/p>\n

                        \u56e0\u679c+X\uff1a\u89e3\u51b3\u591a\u5b66\u79d1\u9886\u57df\u7684\u56e0\u679c\u95ee\u9898 | \u56e0\u679c\u79d1\u5b66\u8bfb\u4e66\u4f1a\u7b2c\u4e09\u5b63\u542f\u52a8<\/a><\/span><\/strong><\/span><\/p>\n

                        \n


                        <\/p>\n<\/section>\n


                        <\/p>\n

                        \n

                        \u63a8\u8350\u9605\u8bfb<\/span><\/strong><\/strong><\/p>\n

                        1.  <\/strong>\u9762\u5411\u56e0\u679c\u89c4\u5f8b\u7684\u8868\u793a\u5b66\u4e60\u65b0\u65b9\u6cd5\u2014\u2014\u56e0\u679c\u8868\u5f81\u5b66\u4e60\u6700\u65b0\u653b\u7565<\/a><\/strong><\/span><\/strong><\/strong><\/section>\n
                        2. <\/strong>\u5468\u6653\u534e\uff1a\u5982\u4f55\u7528\u6570\u5b66\u8bed\u8a00\u6765\u63cf\u8ff0\u8fdb\u884c\u56e0\u679c\u5206\u6790\uff1f<\/a><\/strong><\/span><\/strong><\/strong><\/section>\n
                        3. <\/strong>\u4e01\u9e4f\uff1a\u591a\u89d2\u5ea6\u56de\u987e\u56e0\u679c\u63a8\u65ad\u7684\u6a21\u578b\u65b9\u6cd5<\/a><\/strong><\/span><\/section>\n
                        4. <\/strong><\/span>\u5f20\u6c5f\uff1a\u7b2c\u4e09\u4ee3\u4eba\u5de5\u667a\u80fd\u6280\u672f\u57fa\u7840\u2014\u2014\u4ece\u53ef\u5fae\u5206\u7f16\u7a0b\u5230\u56e0\u679c\u63a8\u7406 | \u96c6\u667a\u5b66\u56ed\u5168\u65b0\u8bfe\u7a0b<\/a><\/strong><\/strong><\/span><\/section>\n
                        5. \u6210\u4e3a\u96c6\u667aVIP\uff0c\u89e3\u9501\u5168\u7ad9\u8bfe\u7a0b\uff0f\u8bfb\u4e66\u4f1a<\/a><\/span><\/strong><\/strong><\/strong><\/section>\n
                        6. <\/strong>\u52a0\u5165\u96c6\u667a\uff0c\u4e00\u8d77\u590d\u6742\uff01<\/strong><\/span><\/a><\/section>\n


                        <\/p>\n


                        <\/p>\n<\/section>\n

                        \u70b9\u51fb\u201c\u9605\u8bfb\u539f\u6587\u201d\uff0c\u62a5\u540d\u8bfb\u4e66\u4f1a<\/span><\/strong><\/section>\n

                        <\/mp-style-type><\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

                        \u5bfc\u8bed \u56e0\u679c\u63a8\u65ad\u7684\u76ee\u6807\u662f\u5c06\u5916\u90e8\u77e5\u8bc6\u4e0e\u5b66\u4e60\u8bbe\u8ba1\u7ed3\u5408\u8d77\u6765\uff0c\u5f97\u5230\u53d8\u91cf\u4e4b\u95f4\u7684\u56e0\u679c\u5173\u7cfb\u3002\u5c24\u5176\u662f\u5728\u4eba\u5de5\u667a\u80fd\u6280\u672f\u906d\u9047\u74f6\u9888\u7684\u73b0\u5728\uff0c\u56e0\u679c\u63a8\u7406\u7684\u65b9\u6cd5\u4e5f\u8d8a\u6765\u8d8a\u53d7\u5230\u4e0d\u540c\u9886\u57df\u7684\u91cd\u89c6\uff0c\u5b83\u5728\u4eba\u5de5\u667a\u80fd\u3001\u7edf\u8ba1\u5b66\u3001\u751f\u7269\u7edf\u8ba1\u5b66\u3001\u751f\u7269\u533b\u5b66\u3001\u8ba1\u7b97\u673a\u79d1\u5b66\u3001\u7ecf\u6d4e\u5b66\u3001\u6d41\u884c\u75c5\u5b66\u548c\u5404\u79cd\u793e\u4f1a\u79d1\u5b66\u7b49\u9886\u57df\u83b7\u5f97\u4e86\u5e7f\u6cdb\u7684\u5e94\u7528\u3002 \u4e3a\u4e86\u66f4\u591a\u5730\u63a8\u52a8\u56e0\u679c\u79d1\u5b66\u5b66\u79d1\u7684\u53d1\u5c55\uff0c…<\/p>\n","protected":false},"author":0,"featured_media":46263,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"special":[],"_links":{"self":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/46266"}],"collection":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=46266"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/46266\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/46263"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=46266"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=46266"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=46266"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=46266"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}