A Tutorial on Logistic Regression Ying So, SAS Institute Inc., Cary, NC

lump together and identify various portions of an otherwise continuous variable, Let T be the underlying continuous variable and suppose that

ABSTRACT Many procedures in SAs/STAT~ can be used to perform 10' gistic regression analysis: CATMOD, GENMOD,LOGISTIC, al)d PROBIT. Each procedure has special features that make. ~ useful for gertain applications. For most applica· tions, PROC LOGISTIC is the preferred choice,. It fits binary response or proportional odds models, provides various model·selection methods to identify important prognostic variables from a large number of candidate variables, and computes regression diagnostic statistics. This tutorial dis· cusses some of the problems users encountered when they used the LOGISTIC procedure.

Y = r

ff