Maximum likelihood estimates for the logit model and the iterative scaling method
- 1 January 1974
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics
- Vol. 3 (10) , 985-993
- https://doi.org/10.1080/03610917408827200
Abstract
Suppose that we have k 2 × c contingency tables having c independent binomial populations b(n:ij:, p:ij:). It is proved that the likelihood equation for the parameters of the logit model for p:ij: has the unique solution, if it is “consistent”. The unique solution attains the true maximum of the likelihood and the existence of the solution is provided by the iterative scaling method due to Deming and Stephan (1940) and Ireland and Kullback (1968), The methods of proof are based on Darroch and Ratcliff (1972).Keywords
This publication has 5 references indexed in Scilit:
- Maximum likelihood estimates for the logit model and the iterative scaling methodCommunications in Statistics, 1974
- Linear Statistical Inference and its ApplicationsPublished by Wiley ,1973
- Generalized Iterative Scaling for Log-Linear ModelsThe Annals of Mathematical Statistics, 1972
- Contingency tables with given marginalsBiometrika, 1968
- On a Least Squares Adjustment of a Sampled Frequency Table When the Expected Marginal Totals are KnownThe Annals of Mathematical Statistics, 1940