STA 290 Seminar Series
DATE: Thursday, April 6th 2017, 4:10pm
LOCATION: MSB 1147, Colloquium Room. Refreshments at 3:30pm in MSB 4110
SPEAKER: Heping Zhang, Susan Dwight Bliss Professor of Biostatistics / Professor of Statistics and Child Study, Yale University
TITLE: “Surrogate residuals: a little thing to pick up”
ABSTRACT: Residual diagnostics in regression is an important classroom topic in statistics. These days, however, it only appears once in a while in a statistical journal, and is hardly mentioned in research publications where regression or association analyses of real data are critical to the findings and conclusions. Perhaps it is too simple, even trivial, to be worth these mentions. Or, perhaps, such perceptions keep us from admitting that we don't know how to do residual diagnostics, or get away from it, such as in the case of logistic regression. This may be even more so when the outcome of interest is an ordinal category such as how sad, good or great we feel today. In this talk, I will attempt to draw your attention to this classic classroom topic, which in my own experience remains very important yet insufficiently attended in real data applications. Using a combination of classic and contemporary statistical techniques, I will introduce a concept of surrogate residuals that appears informative and useful in residual diagnosis of regression models involving categorical outcomes, including logistic regression. This will be supported in theory, simulation, and real data analysis. The work is a collaboration with Dr. Dungang Liu, Assistant Professor of Business Analytics, Carl H. Lindner College of Business, University of Cincinnati.