Bootstrapping Hausman's exogeneity test

Ka-Fu, Wong

doi:10.1016/s0165-1765(96)00917-2

Cited by 21 publications

(20 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bootstrap routines for testing the orthogonality of all possibly endogenous regressors have previously been discussed by Wong (1996). Implementation of these bootstrap routines is relatively easy due to the fact that no regressors are assumed to be endogenous under the null hypothesis.…”

Section: A Bootstrap Routine For Sub-set Dwh Test Statisticsmentioning

confidence: 99%

The performance of tests on endogeneity of subsets of explanatory variables scanned by simulation

Kiviet

Pleus²

2017

Econometrics and Statistics

View full text Add to dashboard Cite

Section: A Bootstrap Routine For Sub-set Dwh Test Statisticsmentioning

confidence: 99%

The performance of tests on endogeneity of subsets of explanatory variables scanned by simulation

Kiviet

Pleus²

2017

Econometrics and Statistics

View full text Add to dashboard Cite

“…Wong (1996) illustrates through a Monte Carlo experiment that bootstrapping the Hausman (1978) exogeneity test improves both the size and power of the test. Li (2006) extends Wong's (1996) results by allowing for serial correlated errors. Both papers are referenced in the weak instrument literature.…”

Section: Bootstrap Validity For Dwh Testsmentioning

confidence: 99%

“…Both papers are referenced in the weak instrument literature. However, neither Wong (1996) nor Li (2006) provides a formal proof of the large-sample validity of their bootstrap, even when IVs are strong. Furthermore, the Monte Carlo designs in both papers exclude cases where IVs are poor, because the smallest correlation between each IV and the (possibly) endogenous regressors is set at 0.1.…”

Section: Bootstrap Validity For Dwh Testsmentioning

confidence: 99%

“…We mainly focus on linear structural models and establish the asymptotic validity of the bootstrap for DWH exogeneity tests, when IVs may be arbitrary weak (weak instruments). Moreira and Suarez (2009) show in the context of hypotheses specified on struc-1 See, for example, Durbin (1954), Wu (1973Wu ( , 1974Wu ( , 1983aWu ( , 1983b, Revankar and Hartley (1973), Farebrother (1976), Hausman (1978), Revankar (1978), Dufour (1979Dufour ( , 1987, Hwang (1980), Kariya and Hodoshima (1980), Hausman and Taylor (1981), Spencer and Berk (1981), Nakamura and Nakamura (1981), Engle (1982), Holly (1982), Reynolds (1982), Smith (1983Smith ( , 1984, Thurman (1986), Smith and Pesaran (1990), Ruud (1984Ruud ( , 2000, Newey (1985aNewey ( , 1985b, Wong (1996), Ahn (1997), Baum, Schaffer and Stillman (2003).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On bootstrap validity for specification tests with weak instruments

Tchatoka

2015

The Econometrics Journal

View full text Add to dashboard Cite

This paper investigates the asymptotic validity of the bootstrap for Durbin-WuHausman (DWH) specification tests when instrumental variables (IVs) may be arbitrary weak. It is shown that under strong identification, the bootstrap offers a better approximation than the usual asymptotic χ 2 distributions. However, the bootstrap provides only a first-order approximation when instruments are weak. These resultsshow unlike the Wald-statistic based on a k-class type estimator (Moreira et al., 2009), the bootstrap is valid even for the Wald-type of DWH statistics in the presence of weak instruments.

show abstract

“…Staiger and Stock (1994) show that 'weak' instruments weaken the power of Hausman test. Wong (1996) proposes a bootstrap procedure to improve the finite sample properties of Hausman test when the instruments are weak. This paper identifies another potential problem of Hausman test.…”

Section: Introductionmentioning

confidence: 99%