光華講壇——社會(huì)名流與企業(yè)家論壇第6612期
主 題：Principal Stratification with Continuous Post-Treatment Variables:Nonparametric Identification and Semiparametric Estimation帶有連續(xù)治療后變量的主要分層：非參數(shù)識(shí)別和半?yún)?shù)估計(jì)
主講人：加州大學(xué)伯克利分校 丁鵬副教授
主持人：統(tǒng)計(jì)學(xué)院林華珍教授
時(shí)間：7月16日 16:00-17:00
舉辦地點(diǎn)：柳林校區(qū)弘遠(yuǎn)樓 408 會(huì)議室
主辦單位：統(tǒng)計(jì)研究中心和統(tǒng)計(jì)學(xué)院 科研處
主講人簡(jiǎn)介：
Peng Ding is an Associate Professor in the Department of Statistics at UC Berkeley. He obtained his Ph.D. from the Department of Statistics, Harvard University in May 2015, and worked as a postdoctoral researcher in the Department of Epidemiology, Harvard T. H. Chan School of Public Health until December 2015. Previously, he received my B.S. in Mathematics,B.A. in Economics, and M.S. in Statistics from Peking University.
丁鵬，加州大學(xué)伯克利分校統(tǒng)計(jì)系的副教授。他于2015年5月在哈佛大學(xué)統(tǒng)計(jì)系獲得博士學(xué)位，并在2015年12月之前在哈佛大學(xué)陳曾熙公共衛(wèi)生學(xué)院流行病學(xué)系擔(dān)任博士后研究員。在此之前，他獲得了北京大學(xué)的數(shù)學(xué)學(xué)士學(xué)位、經(jīng)濟(jì)學(xué)學(xué)士學(xué)位和統(tǒng)計(jì)學(xué)碩士學(xué)位。
內(nèi)容簡(jiǎn)介：
Post-treatment variables often complicate causal inference. They appear in many scientific problems, including noncompliance, truncation by death, mediation, and surrogate endpoint evaluation. Principal stratification is a strategy to address these challenges by adjusting for the potential values of the post-treatment variables, defined as the principal strata. It allows for characterizing treatment effect heterogeneity across principal strata and unveiling the mechanism of the treatment's impact on the outcome related to post-treatment variables.However, the existing literature has primarily focused on binary post-treatment variables, leaving the case with continuous post-treatment variables largely unexplored. This gap persists due to the complexity of infinitely many principal strata, which present challenges to both the identification and estimation of causal effects. We fill this gap by providing nonparametric identification and semiparametric estimation theory for principal stratification with continuous post-treatment variables. We propose to use working models to approximate the underlying causal effect surfaces and derive the efficient influence functions of the corresponding model parameters. Based on the theory, we construct doubly robust estimators and implement them in an R package.
治療后變量通常會(huì)使因果推斷變得復(fù)雜。它們出現(xiàn)在許多科學(xué)問(wèn)題中，包括不遵從、死亡截?cái)?、中介效?yīng)和替代終點(diǎn)評(píng)估。主要分層是一種通過(guò)調(diào)整治療后變量的潛在值（即主要分層）來(lái)解決這些挑戰(zhàn)的策略。它允許表征不同主要分層中的治療效果異質(zhì)性，并揭示治療對(duì)與治療后變量相關(guān)的結(jié)果的影響機(jī)制。然而，現(xiàn)有文獻(xiàn)主要集中在二元治療后變量上，對(duì)于連續(xù)治療后變量的情況則研究較少。由于無(wú)限多的主要分層的復(fù)雜性，這一領(lǐng)域在因果效應(yīng)的識(shí)別和估計(jì)方面面臨挑戰(zhàn)。主講人通過(guò)提供連續(xù)治療后變量主要分層的非參數(shù)識(shí)別和半?yún)?shù)估計(jì)理論填補(bǔ)了這一空白。主講人提出使用工作模型來(lái)逼近潛在的因果效應(yīng)面，并推導(dǎo)出相應(yīng)模型參數(shù)的有效影響函數(shù)。基于該理論，主講人構(gòu)建雙重穩(wěn)健估計(jì)量，并在 R 軟件包中實(shí)現(xiàn)這些方法。
