Testing and Adjusting for Departures from Nominal Dispersion in Generalized Linear Models

SUMMARY In this paper we describe a score test of the hypothesis of no departure from nominal dispersion in a generalized linear model. We also give a method for adjusting the nominal variance-covariance matrix of the estimated regression coefficients when overdispersion is suspected. This procedure is an alternative to the traditional method of adjusting each element of the variance-covariance matrix by the same factor. We illustrate our method for the one-parameter exponential family of distributions in a logistic analysis of a factorial experiment and a Poisson regression analysis of bioassay data.

[1]  D. Cox,et al.  Analysis of Binary Data (2nd ed.). , 1990 .

[2]  J. Neyman,et al.  LOCALLY OPTIMAL TEST FOR HOMOGENEITY WITH RESPECT TO VERY RARE EVENTS. , 1965, Proceedings of the National Academy of Sciences of the United States of America.

[3]  N. Breslow.  Extra‐Poisson Variation in Log‐Linear Models , 1984 .

[4]  Rajesh Kumar. Barnwal.  Analysis of one-way layout of count data. , 1988 .

[5]  P. Armitage.  Host Variability in Dilution Experiments , 1959 .

[6]  Daniel Zelterman,et al.  Homogeneity Tests against Central-Mixture Alternatives , 1988 .

[7]  B. G. Lindsay,et al.  Nuisance parameters, mixture models, and the efficiency of partial likelihood estimators , 1980, Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences.

[8]  B. Efron.  Double Exponential Families and Their Use in Generalized Linear Regression , 1986 .

[9]  K. Mccarthy,et al.  The antibody response in man following infection with viruses of the pox group I. An evaluation of the pock counting method for measuring neutralizing antibody , 1958, Journal of Hygiene.

[10]  H. White.  Maximum Likelihood Estimation of Misspecified Models , 1982 .

[11]  R F Potthoff,et al.  Testing for homogeneity. II. The Poisson distribution. , 1966, Biometrika.

[12]  Rory A. Fisher,et al.  The significance of deviations from expectation in a Poisson Series. , 1950 .

[13]  R. Royall.  Model robust confidence intervals using maximum likelihood estimators , 1986 .

[14]  P. Moran.  The dilution assay of viruses , 1954, Journal of Hygiene.

[15]  Ian Diamond,et al.  Analysis of Binary Data. 2nd Edn. , 1990 .

[16]  David R. Cox,et al.  Some remarks on overdispersion , 1983 .

[17]  J. Lawless,et al.  Tests for Detecting Overdispersion in Poisson Regression Models , 1989 .

[18]  Dirk F. Moore,et al.  Asymptotic properties of moment estimators for overdispersed counts and proportions , 1986 .

[19]  R. Woolson,et al.  A Monte Carlo investigation of homogeneity tests of the odds ratio under various sample size configurations. , 1989, Biometrics.

[20]  Jeffrey R. Wilson.  Chi-square tests for overdispersion with multiparameter estimates , 1989 .

[21]  D. A. Williams,et al.  Extra‐Binomial Variation in Logistic Linear Models , 1982 .

[22]  Dorothy A. Anderson.  SOME MODELS FOR OVERDISPERSED BINOMIAL DATA , 1988 .

[23]  Norman L. Johnson,et al.  The Truncated Poisson , 1952 .

[24]  K. Liang.  A locally most powerful test for homogeneity with many strata , 1987 .

[25]  Gordon K. Smyth,et al.  Generalized linear models with varying dispersion , 1989 .

[26]  Martin Crowder,et al.  Beta-binomial Anova for Proportions , 1978 .

[27]  P. Armitage,et al.  The detection of variation in host susceptibility in dilution counting experiments , 1956, Journal of Hygiene.