On testing sample selection bias under the multicollinearity problem

T. Yamagata; C.D. Orme

doi:10.1080/02770900500406132

On testing sample selection bias under the multicollinearity problem

T. Yamagata, C.D. Orme

Economics

Research output: Contribution to journal › Article › peer-review

Abstract

This paper reviews and extends the literature on the finite sample behavior of tests for sample selection bias. Monte Carlo results show that, when the “multicollinearity problem” identified by Nawata (1993) is severe, (i) the t-test based on the Heckman-Greene variance estimator can be unreliable, (ii) the Likelihood Ratio test remains powerful, and (iii) nonnormality can be interpreted as severe sample selection bias by Maximum Likelihood methods, leading to negative Wald statistics. We also confirm previous findings (Leung and Yu, 1996) that the standard regression-based t-test (Heckman, 1979) and the asymptotically efficient Lagrange Multiplier test (Melino, 1982), are robust to nonnormality but have very little power.

Original language	English
Pages (from-to)	467-481
Number of pages	14
Journal	Econometric Reviews
Volume	24
Issue number	4
DOIs	https://doi.org/10.1080/02770900500406132
Publication status	Published - Oct 2005

Access to Document

10.1080/02770900500406132

http://dx.doi.org/10.1080/02770900500406132

Cite this

@article{1d3a1e8f91a042d7bc2413e09a473a60,

title = "On testing sample selection bias under the multicollinearity problem",

abstract = "This paper reviews and extends the literature on the finite sample behavior of tests for sample selection bias. Monte Carlo results show that, when the “multicollinearity problem” identified by Nawata (1993) is severe, (i) the t-test based on the Heckman-Greene variance estimator can be unreliable, (ii) the Likelihood Ratio test remains powerful, and (iii) nonnormality can be interpreted as severe sample selection bias by Maximum Likelihood methods, leading to negative Wald statistics. We also confirm previous findings (Leung and Yu, 1996) that the standard regression-based t-test (Heckman, 1979) and the asymptotically efficient Lagrange Multiplier test (Melino, 1982), are robust to nonnormality but have very little power.",

author = "T. Yamagata and C.D. Orme",

year = "2005",

month = oct,

doi = "10.1080/02770900500406132",

language = "English",

volume = "24",

pages = "467--481",

journal = "Econometric Reviews",

issn = "0747-4938",

publisher = "Taylor & Francis",

number = "4",

}

TY - JOUR

T1 - On testing sample selection bias under the multicollinearity problem

AU - Yamagata, T.

AU - Orme, C.D.

PY - 2005/10

Y1 - 2005/10

N2 - This paper reviews and extends the literature on the finite sample behavior of tests for sample selection bias. Monte Carlo results show that, when the “multicollinearity problem” identified by Nawata (1993) is severe, (i) the t-test based on the Heckman-Greene variance estimator can be unreliable, (ii) the Likelihood Ratio test remains powerful, and (iii) nonnormality can be interpreted as severe sample selection bias by Maximum Likelihood methods, leading to negative Wald statistics. We also confirm previous findings (Leung and Yu, 1996) that the standard regression-based t-test (Heckman, 1979) and the asymptotically efficient Lagrange Multiplier test (Melino, 1982), are robust to nonnormality but have very little power.

AB - This paper reviews and extends the literature on the finite sample behavior of tests for sample selection bias. Monte Carlo results show that, when the “multicollinearity problem” identified by Nawata (1993) is severe, (i) the t-test based on the Heckman-Greene variance estimator can be unreliable, (ii) the Likelihood Ratio test remains powerful, and (iii) nonnormality can be interpreted as severe sample selection bias by Maximum Likelihood methods, leading to negative Wald statistics. We also confirm previous findings (Leung and Yu, 1996) that the standard regression-based t-test (Heckman, 1979) and the asymptotically efficient Lagrange Multiplier test (Melino, 1982), are robust to nonnormality but have very little power.

U2 - 10.1080/02770900500406132

DO - 10.1080/02770900500406132

M3 - Article

SN - 0747-4938

VL - 24

SP - 467

EP - 481

JO - Econometric Reviews

JF - Econometric Reviews

IS - 4

ER -