Download PDF
Methods Inf Med 1992; 31(03): 215-218
DOI: 10.1055/s-0038-1634875
Original Article
Schattauer GmbH

Maximum Likelihood Estimation and Testing of a Poisson Regression Model

J. Y. Wan
1   Departments of Biostatistics and Geriatric Research and Training Center
,
A. T. Galecki
2   University of Michigan, Ann Arbor, Michigan, USA
› Author Affiliations
Further Information

Publication History

Publication Date:
08 February 2018 (online)

    Permissions and Reprints

Abstract:

A Poisson regression model is proposed for the analysis of incidence rates presented in a two-way table classified by two categorical variables. It is shown that the likelihood function is the same as that using Glasser’s exponential covariate model. An algorithm is given to solve the maximum likelihood estimates of the regression parameters. The model is evaluated via deviance and the method is illustrated with an example. Some extensions of the model are discussed.

Key-Words

Deviance - Exponential Regression - Incidence Rate - Poisson Regression
 

  • REFERENCES

  • 1 Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J National Cancer Inst 1959; 22: 719-48.
    PubMedGoogle Scholar
  • 2 Tarone RE. On summary estimators of relative risk. J Chron Dis 1981; 34: 463-8.
    CrossrefPubMedGoogle Scholar
  • 3 Glasser M. Exponential survival with covariance. J Amer Statist Assoc 1967; 62: 561-8.
    CrossrefPubMedGoogle Scholar
  • 4 Prentice RL. Exponential survivals with censoring and explanatory variables. Biometrika 1973; 60: 279-88.
    CrossrefPubMedGoogle Scholar
  • 5 Cramer EM, Appelbaum Ml. Nonorthogonal analysis of variance - once again. Psychol Bull 1980; 87: 51-7.
    CrossrefPubMedGoogle Scholar
  • 6 Laird N, Olivier D. Covariance analysis of censored survival data using log-linear analysis techniques. J Amer Statist Assoc 1981; 76: 231-40.
    CrossrefPubMedGoogle Scholar
  • 7 Holford TR. The analysis of rates and of survivorship using log-linear models. Biometrics 1980; 36: 299-305.
    CrossrefPubMedGoogle Scholar
  • 8 Breslow NE, Day NE. Statistical Methods in Cancer Research: Volume 2 - The Design and Analysis of Cohort Studies. IARC Scientific Publications No. 82. 1987
    PubMedGoogle Scholar
  • 9 Cutler S, Young J. eds. NCI Monograph 41, DHEW No. NIH-75-787. Third National Cancer Survey: Incidence Data. Bethesda Md: National Cancer Institute; 1975
    Google Scholar
  • 10 Gail M. The analysis of heterogeneity for indirect standardized mortaliy ratios. J Royal Statist Soc Ser A 1978; 141: 224-34.
    PubMedGoogle Scholar
  • 11 SAS Institute Inc. SAS Technical Report P-200, SAS/STAT Software: CALIS and LOGISTIC Procedures, Release 6.04. Carry. NC: SAS Institute Inc; 1990
    Google Scholar
  • 12 Bartholomew DJ. The sampling distribution of an estimate arising in the life testing. Technometrics 1963; 05: 361-74.
    CrossrefPubMedGoogle Scholar
  • 13 Armitage P. Statistical Methods in Medical Research. 2nd ed.. Oxford: Blackwell Scientific Publications; 1987
    Google Scholar
  • 14 Frame EL, Checkoway H. Use of Poisson regression models in estimating incidence rates and ratios. Amer J Epidemiol 1985; 121: 309-23.
    CrossrefPubMedGoogle Scholar
  • 15 Rao CR. Linear Statistical Inference and its Applications. 2nd ed.. New York: Wiley; 1973
    Google Scholar