Incorrect asymptotic size of subsampling procedures based on post-consistent model selection estimators

By Donald W. K. Andrews, Patrik Guggenberger. 2009

46 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/7f9562e6c2a9fc5bc0f4488cda85a704e83d3cd8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

The impact of a Hausman pretest on the size of a hypothesis test: The panel data case (2010, background): “…(see e.g.Wooldridge, 2002, p. 290), then the problem considered here could potentially be embedded into the previously studied problem of pretesting whether a regressor variable should be included in a linear regression model, see e.g. Andrews and Guggenberger (2009a) and Leeb and Pötscher (2005).”
Matching and semi-parametric IV estimation, a distance-based measure of migration, and the wages of young men (2011, background): “Specifically, each case they consider where these methods fail involves a test statistic that has a limiting distribution which is discontinuous in a parameter value (see Andrews and Guggenberger 2008a, 2008b, 2009a, 2009b and 2010).”
Statistical Inference in Dynamic Treatment Regimes (2010, methodology): “Pretesting has been used in Econometrics to provide hypothesis tests and confidence intervals in nonregular settings (Andrews 2001, Andrews & Soares 2007; Cheng 2008; Andrews & Guggenberger 2009).”
Bonferroni-Based Size-Correction for Nonstandard Testing Problems (2017, unknown): “…and Pötscher (2005) in the context of inference after model selection, Stoye (2009) in the context of inference on partially identified parameters and Andrews and Guggenberger (2009a) (AG henceforth) in the context of inference on the autoregressive parameter in a first-order autoregressive model.”
THE IMPACT OF A HAUSMAN PRETEST ON THE ASYMPTOTIC SIZE OF A HYPOTHESIS TEST (2009, background): “Note that in Andrews and Guggenberger (2010b, 2009a, 2009b, 2009c, 2010a), the specification for γ has always been chosen such that when γ1 times n r diverges to infinity, the “standard fixed critical value” asymptotic distribution is obtained.”
Generic results for establishing the asymptotic size of confidence sets and tests (2020, background): “The results in Andrews and Guggenberger (2009c) do not cover48 bootstrap or conditional critical values.”
Hybrid and Size-Corrected Subsampling Methods (2007, background): “8Assumption T is not needed in the moment inequality example because subsampling has correct asymptotic size in that example, see Andrews and Guggenberger (2009d).”
VALIDITY OF SUBSAMPLING AND “PLUG-IN ASYMPTOTIC” INFERENCE FOR PARAMETERS DEFINED BY MOMENT INEQUALITIES (2009, methodology): “For the case of PSC-FCV tests, which are relevant here, only Assumptions A, B, L(i), N, and O(a), defined in Andrews and Guggenberger (2009a), are needed for Theorem 3 of Andrews and Guggenberger (2009a).”
Robust Inference on Average Treatment Effects with Possibly More Covariates than Observations (2013, unknown): “This method of proving uniformity follows Belloni, Chernozhukov, and Hansen (2013) and Romano (2004), and is distinct from the approach of Andrews and Guggenberger (2009).”
Bootstrapping Lasso Estimators (2011, background): “Also, as noted by Andrews and Guggenberger (2009), the existence of a uniform consistent estimator of the sampling distribution is not necessary to achieve the goal of producing a uniformly valid confidence interval.”
Dynamic treatment regimes: technical challenges and applications (2014, methodology): “Moving-parameter asymptotic results better reflect small sample behavior because they allow H⊺2,1β ∗ 2,1 to remain near to zero even in infinite samples (Leeb and Pötscher, 2003, 2005; Andrews and Guggenberger, 2009; Laber and Murphy, 2011).”