Mon, March 20, 2017, noon:
Dean Yang, Taken by Storm
Noreen Goldman (Office of Population Research and Woodrow Wilson School of Public and International Affairs, Princeton University)
09/10/2012, at noon in room 6050 ISR-Thompson.
Researchers frequently use statistical models that incorporate social and demographic covariates to predict survival. Based on data from a biosocial survey in Taiwan, I explore the extent to which mortality prediction improves with the inclusion in such models of non-standard biological and clinical parameters; measures of physical performance; and global health ratings made by interviewers and physicians. Some surprising results emerge.