题 目： Sensitivity testing for nonidentifiable models, with application to
longitudinal data with noninformative dropout
报告人： Prof. Jason Fine（University of Wisconsin-Madison）
时 间： 2010年7月13日（周二）下午2：00
地 点： 成人直播新楼215教室
摘 要：I consider the problem of evaluating a statistical hypothesis when some model characteristics are non-identifiable from observed data. Such scenario is common in meta-analysis for assessing publication bias and in longitudinal studies for evaluating a covariate effect when dropouts are likely to be informative. One possible approach to this problem is to fix a minimal set of sensitivity parameters conditional upon which hypothesized parameters are identifiable.I discuss existing approaches to inference derived by assessing the sensitivity of parameter estimates to the sensitivity parameter. I propose to formally evaluate the hypothesis of interest using an infimum statistic over the whole support of the sensitivity parameter, along with the associated inferential challenges. I characterize the limiting distribution of the statistic as a process in the sensitivity parameter, which involves a careful theoretical analysis of its behavior under model misspecification. In practice, I suggest a nonparametric bootstrap procedure to implement this infimum test as well as to construct confidence bands for simultaneous pointwise tests across all values of the sensitivity parameter, adjusting for multiple testing. The methodology's practical utility is illustrated in an analysis of a longitudinal psychiatric study.
报告人简介：Dr. Jason Fine obtained his PHD in biostatistics from Harvard University in the year of 1998. Then, he jointed Department of Statistics & Department of Biostatistics at University of Wisconsin-Madison till 2008. He is current a professor of biostatistics at University of North Carolina Chapel Hill.Prof. Fine is a very distinguished researcher with many fine publications in both statistics and medical journals, e.g., JASA, Biometrika, Biometrics,Statistics in Medicine, Biostatistics, etc. Prof. Fine is Co-Editor for Statistical Advances in Biomedical Sciences. He is in the editorial board ofJournal of the American Academy of Child and Adolescent Psychiatry. He is also Associate Editor for a number of well respected statistical journals, including: Biometrics, Biostatistics, Scandinavian Journal of Statistics, Lifetime Data Analysis, and also JRSS-C.
