2019/07/14 by Gregory A. Cox, Cox, Gregory, Xiaoxia Shi +1
Mathematics · #Advanced Causal Inference Techniques #Econometrics (econ.EM) #FOS: Economics and business #FOS: Mathematics #Statistical Methods and Bayesian Inference #Statistical Methods and Inference #Statistics Theory (math.ST)
paper · pdf · doi:10.48550/arxiv.1907.06317
openalex publication_date 2019/07/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We propose a simple test for moment inequalities that has exact size in\nnormal models with known variance and has uniformly asymptotically exact size\nmore generally. The test compares the quasi-likelihood ratio statistic to a\nchi-squared critical value, where the degree of freedom is the rank of the\ninequalities that are active in finite samples. The test requires no simulation\nand thus is computationally fast and especially suitable for constructing\nconfidence sets for parameters by test inversion. It uses no tuning parameter\nfor moment selection and yet still adapts to the slackness of the moment\ninequalities. Furthermore, we show how the test can be easily adapted for\ninference on subvectors for the common empirical setting of conditional moment\ninequalities with nuisance parameters entering linearly.\n