An Algorithm for Estimation in Simultaneous Logit Models.
Abstract
This paper uses a very general version of the Iterative Proportional Fitting Procedure to develop an algorithm for estimation in simultaneous logit models. The algorithm can be used for any loglinear model which can be cast in the form of simultaneous logit equations. The principal advantage of this method is that it is not necessary to fit parameters associated with the sampling constraints and thus very large problems can be attacked. A numerical example using GLIM and a sample GLIM macro are included. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1981
- Accession Number
- ADA111015
Entities
People
- Michael M. Meyer
Organizations
- Carnegie Mellon University