Valid Wald Inference with Many Weak Instruments
Econometrics
2023-11-28 v1
Abstract
This paper proposes three novel test procedures that yield valid inference in an environment with many weak instrumental variables (MWIV). It is observed that the t statistic of the jackknife instrumental variable estimator (JIVE) has an asymptotic distribution that is identical to the two-stage-least squares (TSLS) t statistic in the just-identified environment. Consequently, test procedures that were valid for TSLS t are also valid for the JIVE t. Two such procedures, i.e., VtF and conditional Wald, are adapted directly. By exploiting a feature of MWIV environments, a third, more powerful, one-sided VtF-based test procedure can be obtained.
Keywords
Cite
@article{arxiv.2311.15932,
title = {Valid Wald Inference with Many Weak Instruments},
author = {Luther Yap},
journal= {arXiv preprint arXiv:2311.15932},
year = {2023}
}