DMCA
MIXED MNL MODELS FOR DISCRETE RESPONSE (2000)
Cached
Download Links
| Venue: | JOURNAL OF APPLIED ECONOMETRICS J. APPL. ECON. 15: 447--470 (2000) |
| Citations: | 485 - 14 self |
BibTeX
@MISC{Mcfadden00mixedmnl,
author = {Daniel Mcfadden and Kenneth Train},
title = {MIXED MNL MODELS FOR DISCRETE RESPONSE},
year = {2000}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
This paper considers mixed, or random coefficients, multinomial logit (MMNL) models for discrete response, and establishes the following results. Under mild regularity conditions, any discrete choice model derived from random utility maximization has choice probabilities that can be approximated as closely as one pleases by a MMNL model. Practical estimation of a parametric mixing family can be carried out by Maximum Simulated Likelihood Estimation or Method of Simulated Moments, and easily computed instruments are provided that make the latter procedure fairly efficient. The adequacy of a mixing specification can be tested simply as an omitted variable test with appropriately defined artificial variables. An application to a problem of demand for alternative vehicles shows that MMNL provides a flexible and computationally practical approach to discrete response analysis.
Keyphrases
journal applied econometrics mixed mnl model discrete response mixing speci cation following result mmnl model choice probability john wiley son cial variable practical estimation multinomial logit maximum simulated likelihood estimation simulated moment latter procedure random utility maximization variable test mild regularity condition alternative vehicle practical approach discrete choice model response analysis coe cients discrete response
