Graphical Methods for Binary Data

Sanford Weisberg

Department of Applied Statistics, School of Statistics, University of Minnesota

In simple regression, the usual graph of the response versus the predictor contains virtually all the information available about the regression problem, and it therefore provides a good summary of the problem. In regression with a binary response, this graph is not particularly useful becuase the response is only a zero or a one. After a brief review of binary regression basics, the relationship between graphs and logistic regression models will be derived and discussed. More general ideas for exploring binary regression through graphics are then presented.

The final part of the talk covers model checking. In linear models, model checking graphs are based on the fundamental result that given a correct model, residuals and the predictors are uncorrelated, but they are generally not independent for an incorrectly specified model. In binary regression, this property does not hold. so exactly how residuals are to be used in model checking is unclear. Alternative plots called marginal model plots that do not use residuals will be described and discussed.


Refreshments: 3:30 - 4:00 p.m. Tuesday, May 5, at 327 Yost
Talk: 4:00 - 5:00 p.m. Tuesday, May 5, at 327 Yost.

Questions? jiayang@sun.cwru.edu
Wed Aug 13 13:54:29 EDT 1997