Partially Adaptive Estimation of the Censored Regression Model

Abstract

Data censoring causes ordinary least squares estimates of linear models to be biased and inconsistent. Tobit, semiparametric, and partially adaptive estimators have been considered as possible solutions. This paper proposes several new partially adaptive estimators that cover a wide range of distributional characteristics. A simulation study is used to investigate the estimators’ relative efficiency in these settings. The partially adaptive censored regression estimators have little efficiency loss for censored normal errors and may outperform Tobit and semiparametric estimators considered for non-normal distributions. An empirical example of out-of-pocket expenditures for a health insurance plan provides an example, which supports these results.

Keywords:

• Censored regression
• CLAD
• Partially adaptive estimators
• SCLS
• Tobit

JEL Classification:

• C13
• C16
• C24

ACKNOWLEDGMENT

The authors appreciate James Cardon for providing the data and for his comments on the paper and Jason Cook and Jeff Sorensen for their research assistance. The authors would also like to thank participants at the MIT Workshop for comments on an earlier version of this paper.

Notes

N = 200, T = 1, 000 Simulations

N = 1000, T = 1000 simulations.

N = 200, T = 1,000 simulations.

N = 1, 000, T = 1, 000 simulations.

N = 200, T = 1, 000 simulations.

N = 1, 000, T = 1, 000 simulations.

Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/lecr