论文信息 - On the existence of maximum likelihood estimates in logistic regression models

On the existence of maximum likelihood estimates in logistic regression models

SUMMARY The problems of existence, uniqueness and location of maximum likelihood estimates in log linear models have received special attention in the literature (Haberman, 1974, Chapter 2; Wedderburn, 1976; Silvapulle, 1981). For multinomial logistic regression models, we prove existence theorems by considering the possible patterns of data points, which fall into three mutually exclusive and exhaustive categories: complete separation, quasicomplete separation and overlap. Our results suggest general rules for identifying infinite parameter estimates in log linear models for frequency tables.

A. Albert | Jacquelyn L. Anderson

[1] Day Ne,et al. A GENERAL MAXIMUM LIKELIHOOD DISCRIMINANT , 1967 .

[2] David R. Cox. The analysis of binary data , 1970 .

[3] David R. Cox,et al. Regression models and life tables (with discussion , 1972 .

[4] J. Anderson. Separate sample logistic discrimination , 1972 .

[5] J. A. Anderson,et al. Diagnosis by Logistic Discriminant Function: Further Practical Problems and Results , 1974 .

[6] S. Haberman,et al. The analysis of frequency data , 1974 .

[7] R. W. Wedderburn,et al. On the existence and uniqueness of the maximum likelihood estimates for certain generalized linear models , 1976 .

[8] R. Pyke,et al. Logistic disease incidence models and case-control studies , 1979 .

[9] M. Silvapulle. On the Existence of Maximum Likelihood Estimators for the Binomial Response Models , 1981 .

[10] J. A. Anderson,et al. Probit and logistic discriminant functions , 1981 .

[11] J. Anderson,et al. Regression, Discrimination and Measurement Models for Ordered Categorical Variables , 1981 .

[12] J. Anderson,et al. Penalized maximum likelihood estimation in logistic regression and discrimination , 1982 .