報告題目：Regression analysis of group-tested current status data
報 告 人：李樹威 副教授 廣州大學
報告時間：2021年6月25日 9:30-10:30
報告地點：騰訊會議ID：998 864 608
或點擊鍊接直接加入會議：https://meeting.tencent.com/s/A93VzeyexuEK
校内聯系人：趙世舜 zhaoss@jlu.edu.cn
報告摘要：Group testing or pooled testing has been shown to be an effective way to reduce the efforts and costs in large-scale screening studies involving infectious disease. In such applications, instead of testing the infection status based on the sample of each individual, group testing involves pooling samples of several individuals and testing each pooled sample for the presence of the infection. When the endpoint of interest is a time-to-event outcome and the infection status of the pooled individuals is examined once through a test with imperfect sensitivity and specificity, we have group-tested current status data. In this paper, we investigate regression analysis of such data with the commonly used semiparametric proportional hazards model. For the estimation, we present a sieve maximum likelihood approach and develop an expectation-maximization (EM) algorithm to estimate the unknown parameters. The numerical results in the simulation study show the proposed method performs well, and has some desirable advantages over the estimation method based on the individual testing results, such as giving more accurate estimation with much less testing costs. An application to a medical study on chlamydia is also provided.
報告人簡介：李樹威，廣州大學統計系副教授，研究生導師，研究領域為生物統計、生存分析、縱向數據等。目前已發表10餘篇高水平科研論文，主持國家自然科學基金1項。
