2016/11/25 (Fri) 10:20-11:30 Regression models for dynamic risk prediction(File download)

Spearker:Prof. Chung-Chou H. Chang

Introduction:Professor of Medicine, Biostatistics, and Clinical and Translational Science , Department of Medicine , University of Pittsburgh

Time & Place:2016/11/25(Fri.)    3F, Assembly Building I

10:20-10:40 Tea Party(4F-428, Assembly Building I, NCTU)
10:40-11:30 Lectures  (3F-308, Assembly Building I, NCTU)

subject:Regression models for dynamic risk prediction


 Risk prediction modeling has been widely used for assessing the effects of change in risk factors on the absolute risk of disease incidence or disease progression, weighing the risk and benefits of an intervention, and designing future prevention trials, yet most of such models to date are applicable only to information observed at or before study baseline. The recent dynamic risk prediction models incorporating information after study baseline into modeling are more capable of handling the effects of disease progression. To further advance such modeling approach, we proposed a risk prediction model that incorporates longitudinally updated information to account for the effect of competing risks. In this talk, I will review the currently available models, introduce our model, and demonstrate the use of these models with a data set from a multicenter clinical trial for breast cancer patients.

File download:Regression models for dynamic risk-prediction.pdf

Big Data Research Center, NCTU
Department of Applied Mathematics, NCTU
Institute of Statistics, NCTU
Institute of Statistics, NTHU