Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model
In a recent study, it was demonstrated that the maximum simulated likelihood (MSL) estimator produces significant biases when applied to the bivariate normal and bivariate Poisson-lognormal models. The study’s conclusion suggests that similar biases could be present in other models generated by corr...
| Published in: | Econometrics |
|---|---|
| Main Authors: | , , , |
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Subjects: | |
| Online Access: | https://www.mdpi.com/2225-1146/12/2/8 |
| _version_ | 1850102373591023616 |
|---|---|
| author | Maksat Jumamyradov Murat Munkin William H. Greene Benjamin M. Craig |
| author_facet | Maksat Jumamyradov Murat Munkin William H. Greene Benjamin M. Craig |
| author_sort | Maksat Jumamyradov |
| collection | DOAJ |
| container_title | Econometrics |
| description | In a recent study, it was demonstrated that the maximum simulated likelihood (MSL) estimator produces significant biases when applied to the bivariate normal and bivariate Poisson-lognormal models. The study’s conclusion suggests that similar biases could be present in other models generated by correlated bivariate normal structures, which include several commonly used specifications of the mixed logit (MIXL) models. This paper conducts a simulation study analyzing the MSL estimation of the error components (EC) MIXL. We find that the MSL estimator produces significant biases in the estimated parameters. The problem becomes worse when the true value of the variance parameter is small and the correlation parameter is large in magnitude. In some cases, the biases in the estimated marginal effects are as large as 12% of the true values. These biases are largely invariant to increases in the number of Halton draws. |
| format | Article |
| id | doaj-art-ce32c834e2e64f0b97eddb3fea5c2838 |
| institution | Directory of Open Access Journals |
| issn | 2225-1146 |
| language | English |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| spelling | doaj-art-ce32c834e2e64f0b97eddb3fea5c28382025-08-20T00:04:00ZengMDPI AGEconometrics2225-11462024-03-01122810.3390/econometrics12020008Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit ModelMaksat Jumamyradov0Murat Munkin1William H. Greene2Benjamin M. Craig3Department of Economics, University of South Florida, Tampa, FL 33620, USADepartment of Economics, University of South Florida, Tampa, FL 33620, USADepartment of Economics, University of South Florida, Tampa, FL 33620, USADepartment of Economics, University of South Florida, Tampa, FL 33620, USAIn a recent study, it was demonstrated that the maximum simulated likelihood (MSL) estimator produces significant biases when applied to the bivariate normal and bivariate Poisson-lognormal models. The study’s conclusion suggests that similar biases could be present in other models generated by correlated bivariate normal structures, which include several commonly used specifications of the mixed logit (MIXL) models. This paper conducts a simulation study analyzing the MSL estimation of the error components (EC) MIXL. We find that the MSL estimator produces significant biases in the estimated parameters. The problem becomes worse when the true value of the variance parameter is small and the correlation parameter is large in magnitude. In some cases, the biases in the estimated marginal effects are as large as 12% of the true values. These biases are largely invariant to increases in the number of Halton draws.https://www.mdpi.com/2225-1146/12/2/8maximum simulated likelihoodmixed logitdiscrete choice models |
| spellingShingle | Maksat Jumamyradov Murat Munkin William H. Greene Benjamin M. Craig Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model maximum simulated likelihood mixed logit discrete choice models |
| title | Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model |
| title_full | Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model |
| title_fullStr | Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model |
| title_full_unstemmed | Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model |
| title_short | Biases in the Maximum Simulated Likelihood Estimation of the Mixed Logit Model |
| title_sort | biases in the maximum simulated likelihood estimation of the mixed logit model |
| topic | maximum simulated likelihood mixed logit discrete choice models |
| url | https://www.mdpi.com/2225-1146/12/2/8 |
| work_keys_str_mv | AT maksatjumamyradov biasesinthemaximumsimulatedlikelihoodestimationofthemixedlogitmodel AT muratmunkin biasesinthemaximumsimulatedlikelihoodestimationofthemixedlogitmodel AT williamhgreene biasesinthemaximumsimulatedlikelihoodestimationofthemixedlogitmodel AT benjaminmcraig biasesinthemaximumsimulatedlikelihoodestimationofthemixedlogitmodel |