TY - JOUR
T1 - Semiparametric tests of conditional moment restrictions under weak or partial identification
AU - Jun, Sung Jae
AU - Pinkse, Joris
PY - 2009/9/1
Y1 - 2009/9/1
N2 - We propose two new semiparametric specification tests which test whether a vector of conditional moment conditions is satisfied for any vector of parameter values θ0. Unlike most existing tests, our tests are asymptotically valid under weak and/or partial identification and can accommodate discontinuities in the conditional moment functions. Our tests are moreover consistent provided that identification is not too weak. We do not require the availability of a consistent first step estimator. Like Robinson [Robinson, Peter M., 1987. Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form. Econometrica 55, 875-891] and many others in similar problems subsequently, we use k-nearest neighbor (knn) weights instead of kernel weights. The advantage of using knn weights is that local power is invariant to transformations of the instruments and that under strong point identification computation of the test statistic yields an efficient estimator of θ0 as a byproduct.
AB - We propose two new semiparametric specification tests which test whether a vector of conditional moment conditions is satisfied for any vector of parameter values θ0. Unlike most existing tests, our tests are asymptotically valid under weak and/or partial identification and can accommodate discontinuities in the conditional moment functions. Our tests are moreover consistent provided that identification is not too weak. We do not require the availability of a consistent first step estimator. Like Robinson [Robinson, Peter M., 1987. Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form. Econometrica 55, 875-891] and many others in similar problems subsequently, we use k-nearest neighbor (knn) weights instead of kernel weights. The advantage of using knn weights is that local power is invariant to transformations of the instruments and that under strong point identification computation of the test statistic yields an efficient estimator of θ0 as a byproduct.
UR - http://www.scopus.com/inward/record.url?scp=68649087117&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=68649087117&partnerID=8YFLogxK
U2 - 10.1016/j.jeconom.2009.02.005
DO - 10.1016/j.jeconom.2009.02.005
M3 - Article
AN - SCOPUS:68649087117
SN - 0304-4076
VL - 152
SP - 3
EP - 18
JO - Journal of Econometrics
JF - Journal of Econometrics
IS - 1
ER -