Introduction Regression diagnostics are techniques for the detection
and assessment of potential problems resulting from a
fitted regression model that might either support, compro-
mise, or negate the assumptions made about the regres-
sion model and/or the conclusions drawn from the analysis
of one’s data.
In this chapter, we focus on one important issue for evalu-
ating binary logistic regression results, namely, goodness
of fit (GOF) measurement. Although examination of data
for potential problems, such as GOF, has always been con-
sidered a requirement of the analysis, the availability of
computer software to efficiently perform the complex cal-
culations required has contributed greatly to fine-tuning
the diagnostic procedures and the conclusions drawn from
them.Abbreviated Outline
Detailed Outline
The outline below gives the user a preview of the material
to be covered by the presentation. A detailed outline for
review purposes follows the presentation.I. Overview (pages 304–305)
II. Saturated vs. Fully Parameterized Models (pages
305–312)
III. The Deviance Statistic (pages 312–317)
IV. The HL Statistic (pages 318–320)
V. Examples of the HL Statistic (pages 320–325)
VI. Summary (page 326)
VII. Appendix: Derivation of SS Deviance Formula
(pages 327–328)302 9. Assessing Goodness of Fit for Logistic Regression