Heteroscedasticity and Distributional Assumptions in the Censored Regression Model

Data censoring causes ordinary least-square estimators of linear models to be biased and inconsistent. The Tobit estimator yields consistent estimators in the presence of data censoring if the errors are normally distributed. However, nonnormality or heteroscedasticity results in the Tobit estimators being inconsistent. Various estimators have been proposed for circumventing the normality assumption. Some of these estimators include censored least absolute deviations (CLAD), symmetrically censored least-square (SCLS), and partially adaptive estimators. CLAD and SCLS will be consistent in the presence of heteroscedasticity; however, SCLS performs poorly in the presence of asymmetric errors. This article extends the partially adaptive estimation approach to accommodate possible heteroscedasticity as well as nonnormality. A simulation study is used to investigate the estimators’ relative performance in these settings. The partially adaptive censored regression estimators have little efficiency loss for censored normal errors and appear to outperform the Tobit and semiparametric estimators for nonnormal error distributions and be less sensitive to the presence of heteroscedasticity. An empirical example is considered, which supports these results.

[1]  Arthur S. Goldberger,et al.  ABNORMAL SELECTION BIAS , 1983 .

[2]  James L. Powell,et al.  Symmetrically Trimmed Least Squares Estimation For Tobit Models , 1986 .

[3]  W. Greene.  ON THE ASYMPTOTIC BIAS OF THE ORDINARY LEAST SQUARES ESTIMATOR OF THE TOBIT MODEL , 1981 .

[4]  James B. McDonald,et al.  A comparison of some robust, adaptive, and partially adaptive estimators of regression models , 1993 .

[5]  P. Schmidt,et al.  An Investigation of the Robustness of the Tobit Estimator to Non-Normality , 1982 .

[6]  W. Greene,et al.  计量经济分析 = Econometric analysis , 2009 .

[7]  James B. McDonald,et al.  A generalization of the beta distribution with applications , 1995 .

[8]  Laurence L. George,et al.  The Statistical Analysis of Failure Time Data , 2003, Technometrics.

[9]  J. Powell,et al.  Least absolute deviations estimation for the censored regression model , 1984 .

[10]  Steven B. Caudill,et al.  A partially adaptive estimator for the censored regression model based on a mixture of normal distributions , 2012, Stat. Methods Appl..

[11]  James B. McDonald,et al.  Partially Adaptive Estimation of Regression Models via the Generalized T Distribution , 1988, Econometric Theory.

[12]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data: Kalbfleisch/The Statistical , 2002 .

[13]  Thanasis Stengos,et al.  Partially Adaptive Estimation Via the Maximum Entropy Densities , 2005 .

[14]  T. Mroz,et al.  The Sensitivity of an Empirical Model of Married Women's Hours of Work to Economic and Statistical Assumptions , 1987 .

[15]  Econo Metrica.  REGRESSION ANALYSIS WHEN THE DEPENDENT VARIABLE IS TRUNCATED NORMAL , 2016 .

[16]  Randall A. Lewis,et al.  Partially Adaptive Estimation of the Censored Regression Model , 2014 .

[17]  Harry J. Paarsch.  A Monte Carlo comparison of estimators for censored regression models , 1984 .

[18]  N. L. Johnson,et al.  Continuous Univariate Distributions. , 1995 .

[19]  K. Nawata.  Robust estimation based on grouped-adjusted data in censored regression models , 1990 .

[20]  J. Tobin.  Estimation of Relationships for Limited Dependent Variables , 1958 .

[21]  B. Hansen.  Autoregressive Conditional Density Estimation , 1994 .

[22]  J. McDonald.  Partially adaptive estimation of ARMA time series models , 1989 .

[23]  Takeshi Amemiya,et al.  Regression Analysis when the Dependent Variable is Truncated Normal , 1973 .

[24]  Tue Gørgens,et al.  Semiparametric Estimation of Single-Index Hazard Functions Without Proportional Hazards , 2006 .

[25]  T. Amemiya.  Tobit models: A survey , 1984 .

[26]  W. N. Street,et al.  Financial Data and the Skewed Generalized T Distribution , 1998 .

[27]  J. Heckman.  Sample selection bias as a specification error , 1979 .

[28]  Richard J. Butler,et al.  Robust and Partially Adaptive Estimation of Regression Models , 1990 .

[29]  Whitney K. Newey,et al.  Specification tests for distributional assumptions in the tobit model , 1987 .

[30]  Jeffrey M. Wooldridge,et al.  Introductory Econometrics: A Modern Approach , 1999 .

[31]  N. L. Johnson,et al.  Systems of frequency curves generated by methods of translation. , 1949, Biometrika.

[32]  S. Cosslett.  Efficient Semiparametric Estimation of Censored and Truncated Regressions via a Smoothed Self‐Consistency Equation , 2004 .

[33]  James B. McDonald,et al.  A comparison of semi-parametric and partially adaptive estimators of the censored regression model with possibly skewed and leptokurtic error distributions , 1996 .