Semiparametric Censored Regression Models
- 1 November 2001
- journal article
- Published by American Economic Association in Journal of Economic Perspectives
- Vol. 15 (4), 29-42
- https://doi.org/10.1257/jep.15.4.29
Abstract
When data are censored, ordinary least squares regression can provide biased coefficient estimates. Maximum likelihood approaches to this problem are valid only if the error distribution is correctly specified, which can be problematic in practice. We review several semiparametric estimators for the censored regression model that do not require parameterization of the error distribution. These estimators are used to examine changes in black-white earnings inequality during the 1960s based on censored tax records. The results show that there was significant earnings convergence among black and white men in the American South after the passage of the 1964 Civil Rights Act.Keywords
This publication has 15 references indexed in Scilit:
- Estimation of Semiparametric Censored Regression Models: An Application to Changes in Black-White Earnings Inequality during the 1960sThe Journal of Human Resources, 1998
- Estimation of some partially specified nonlinear modelsJournal of Econometrics, 1997
- Changes in the U.S. Wage Structure 1963-1987: Application of Quantile RegressionEconometrica, 1994
- Trimmed Lad and Least Squares Estimation of Truncated and Censored Regression Models with Fixed EffectsEconometrica, 1992
- The Sensitivity of an Empirical Model of Married Women's Hours of Work to Economic and Statistical AssumptionsEconometrica, 1987
- An Investigation of the Robustness of the Tobit Estimator to Non-NormalityEconometrica, 1982
- Further evidence on the robustness of the Tobit estimator to heteroskedasticityJournal of Econometrics, 1981
- Sample Selection Bias as a Specification ErrorEconometrica, 1979
- Regression QuantilesEconometrica, 1978
- Nonparametric Estimation from Incomplete ObservationsJournal of the American Statistical Association, 1958