In regression models not only the parameter estimates and signicances of explanatory variables are of interest, but also the degree to which variation in the dependent variable can be explained by covariates. In recent publications an R 2 -measure based on deviance was recommended for Poisson regression models, one of the most frequently used modelling tools in epidemiological studies. However, when sample size is small relative to the number of covariates in the model, simple R 2 -measures may be seriously inated and may need to be adjusted according to the number of covariates in the model. Two new adjustments for the R 2 -measure in Poisson regression models based on deviance residuals are presented and compared by simulation with population values. The proposed measures are also applied to real data sets. c 2000 Elsevier Science B.V. All rights reserved.
[1]
N. Breslow,et al.
Cohort Analysis in Epidemiology
,
1985
.
[2]
F. Windmeijer,et al.
R-Squared Measures for Count Data Regression Models With Applications to Health-Care Utilization
,
1996
.
[3]
Niels Keiding,et al.
Statistical Models Based on Counting Processes
,
1993
.
[4]
Stephen E. Fienberg,et al.
A Celebration of Statistics
,
1985
.
[5]
R Doll,et al.
Mortality of British doctors in relation to smoking: observations on coronary thrombosis.
,
1966,
National Cancer Institute monograph.
[6]
M Schemper,et al.
Explained variation for logistic regression.
,
1996,
Statistics in medicine.
[7]
P. McCullagh,et al.
Generalized Linear Models
,
1972,
Predictive Analytics.