Two-step estimation of semiparametric censored regression models

By Shakeeb Khan, J. Powell. 2001

105 citations recorded by the source corpus.

Root-n-consistent estimators of the regression coecients in the linear censored regression model under conditional quantile restrictions on the error terms were pro- posed by Powell (Journal of Econometrics 25 (1984) 303-325, 32 (1986a) 143-155). While those estimators have desirable asymptotic properties under weak regularity conditions, simulation studies have shown these estimators to exhibit a small sam- ple bias in the opposite direction of the least squares bias for censored data. This paper introduces two-step estimators for these models which minimize convex objec- tive functions, and are designed to overcome thisnite-sample bias. The paper gives regularity conditions under which the proposed two-step estimators are consistent and asymptotically normal; a Monte Carlo study compares thenite sample behavior of the proposed methods with their one-step counterparts. ? 2001 Elsevier Science S.A. All rights reserved.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/963af747cb6a05398033c11b94d6bda21e429b57

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Efficient Semiparametric Estimation of Quantile Treatment Effects (2004, unknown): “…Xi Yi q τ 1I Yi q k argmin q c Q 1N N∑i 1 Tip̂ Xi 7 Yi q τ 1I Yi q O Yi q1 τ τ 1I Yi q1 τ k 9 argmin q c Q 1N N∑i 1 Tip̂ Xi 7 1I Yi q1 τ τ q q1 τ [ Yi q 1I Yi q1 τ 1I Yi q O 9 (34) 17As an example of a two-step estimator that requires this type of trimming, see for instance, Khan and Powell (2001).”
Three-Step Censored Quantile Regression and Extramarital Affairs (2002, methodology): “Analogously, Khan and Powell (2001) proposed using any of the following three methods to perform the rst-stage selection: maximum score estimators of the regression quantile, nonparametric kernel propensity score estimator for h4Xi5, and the nonparametric locally linear conditional quantile…”
Censored regression quantiles with endogenous regressors (2007, methodology): “The implementation of this estimator is as described in, say, Khan and Powell (2001); in short, this estimator is the intercept term q̂i in problem q̂i ĉi !”
An MCMC approach to classical estimation (2003, unknown): “…to that plotted in Figure 1 – it is nonsmooth and highly nonconvex, with numerous local optima, posing a formidable obstacle to the practical use of this extremum estimator; see Buchinsky and Hahn (1998), Buchinsky (1991), Fitzenberger (1997), and Khan and Powell (2001) for related discussions.”