Non-parametric maximum likelihood estimation of censored regression models (Q1083147)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Non-parametric maximum likelihood estimation of censored regression models |
scientific article |
Statements
Non-parametric maximum likelihood estimation of censored regression models (English)
0 references
1986
0 references
This paper presents a consistent estimator of a censored linear regression model which does not require knowledge of the distribution of the error term. The estimator considered here applies \textit{G. Duncan}'s [A robust regression estimator. Dept. Econ. Working Paper, Washington State Univ., Pullman/WA (1982)] suggestion that the likelihood function for the censored regression model be treated as a functional of both the unknown regression vector and the unknown error distribution. Our estimator is the majorizing regression vector for this non-parametric likelihood functional. We find conditions which ensure the consistency of the NPMLE. The paper concludes with the results of Monte Carlo experiments which show the NPMLE to be more efficient than Powell's least absolute deviations (LAD) estimator [\textit{J. L. Powell}, J. Econ. 25, 303-325 (1984; Zbl 0571.62100)], particularly when the fraction of censored observations is large and the sample size is small.
0 references
asymptotic normality
0 references
relative efficiency
0 references
consistent estimator
0 references
censored linear regression model
0 references
majorizing regression vector
0 references
non-parametric likelihood functional
0 references
consistency
0 references
Monte Carlo experiments
0 references
Powell's least absolute deviations
0 references
0 references
0 references
0 references
0 references