报告题目：Risk Predictions with Applications in Medicine & Law
报告人：ProfessorQing Pan
报告时间：2024年5月20日下午4:00-5:00
报告地点：文波智慧教室109
摘要：We will talk about two projects where cutting-edge statistical prediction methods are applied to issues with great importance in the public health & legal society. In the first project, our work focuses on the prediction of the probability of advanced adenoma in colorectal cancer screening. The approach involves non-stationary Poisson process for adenoma and informative screening events, with semi-parametric Cox models correlated by a latent frailty variable. The study employs estimating equations in Huang et al. (2006) for the estimation of coefficients and baseline intensity functions and the borrow-strength method (Huang and Wang, 2004) for the subject-specific frailty value. Marginal predictionmodel and prediction model conditional on frailty terms with or without screening history are compared. When patient’s screening history is available, the frailty prediction model is preferred. However, early drop-out or censoring time will cause under-estimation of subject-specific frailty and lead to inaccuracy of prediction using frailty models, in which case marginal prediction model have better performance. In the second project, the economic loss due to prevailing plaintiffs in Alexander v. Milwaukee, where white male lieutenants were discriminated in promotion to captain are calculated, which involves the length of time they would have been in the higher position had they been treated fairly during the period in which the employer practiced discrimination. This problem is complicated by the fact that one’s eligibility for promotion is subject to termination by retirement and both the promotion and retirement processes may be affected by discriminatory practices. The semi-competing risk data are decomposed into the marginal retirement process and the promotion process conditional on that retirement has not occurred. Predictions for compensation purpose are made by utilizing the expected promotion and retirement probabilities of similarly qualified members of the non-discriminated group. The restricted mean durations of three periods are estimated - the time an employee would be at the lower position, at the higher level and in retirement.
主讲人简介：Prof Pan got her PhD in Biostatistics from University of Michigan in 2007. She served as deputy chair of statistics at George Washington University in Washington DC, and is currently a full professor with tenure at the Department of Biostatistics & Bioinformatics, School of Public Health, GWU. She is also the co-director of the clinical & Population Core at DC CFAR (an HIV research center). She has been PI and co-I/lead statistician in many US NIH, Gates Foundation, FDA, Department of Justice projects with over 70 publications. She served in many US and international review panels, program committees and editorial boards. She won the Gertrude M. Cox award from the American Statistical Association in 2022.
