On the robustness of the corrected least squares (cols) estimator for the tobit model
Brazilian Review of Econometrics
View Archive InfoField | Value | |
Title |
On the robustness of the corrected least squares (cols) estimator for the tobit model
On the robustness of the corrected least squares (cols) estimator for the tobit model |
|
Creator |
Stengos, Thanasis; University of Guelph, Guelph, Ontario.
|
|
Subject |
—
— — — |
|
Description |
We provide an alternative derivation for the simple moment estimator first developed by Greene. Our derivation is based on a Taylor series approximation to the population regression function. Our sampling experiments suggest that this estimator, although in general inconsistent is quite well-behaved and robust to different distributional assumptions regarding the exogenous variables, provided that the latter have not been generated from highly skewed distributions.
We provide an alternative derivation for the simple moment estimator first developed by Greene. Our derivation is based on a Taylor series approximation to the population regression function. Our sampling experiments suggest that this estimator, although in general inconsistent is quite well-behaved and robust to different distributional assumptions regarding the exogenous variables, provided that the latter have not been generated from highly skewed distributions. |
|
Publisher |
Sociedade Brasileira de Econometria
|
|
Date |
1987-11-02
|
|
Type |
info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion — — |
|
Format |
application/pdf
|
|
Identifier |
http://bibliotecadigital.fgv.br/ojs/index.php/bre/article/view/3100
10.12660/bre.v7n21987.3100 |
|
Source |
Brazilian Review of Econometrics; Vol 7, No 2 (1987); 73–88
Brazilian Review of Econometrics; Vol 7, No 2 (1987); 73–88 1980-2447 |
|
Language |
eng
|
|
Relation |
http://bibliotecadigital.fgv.br/ojs/index.php/bre/article/view/3100/1993
|
|