Tests with correct size when instruments can be arbitrarily weak

By Marcelo J. Moreira. 2009

191 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/35316d8ae99111f8cc4ce3c972ee941029c99467

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

THE IMPACT OF A HAUSMAN PRETEST ON THE ASYMPTOTIC SIZE OF A HYPOTHESIS TEST (2009, methodology): “…weak, that is, the strength of the instruments is not bounded away from zero, my recommendation is to use one of the robust testing procedures suggested by Anderson and Rubin (1949), Kleibergen (2002), Moreira (2003, 2009), Guggenberger and Smith (2005), or Andrews, Moreira, and Stock (2006).”
Bootstrap inference in a linear equation estimated by instrumental variables (2008, unknown): “We echo Moreira et al. (2007), however, in noting that it is remarkable that the bootstrap Wald and LR tests perform as well as they do.”
ON THE ASYMPTOTIC SIZE DISTORTION OF TESTS WHEN INSTRUMENTS LOCALLY VIOLATE THE EXOGENEITY ASSUMPTION (2011, unknown): “The CUE was introduced by Hansen, Heaton, and Yaron (1996), EL by Owen (1988), Qin and Lawless (1994), Imbens (1997), and Kitamura (2001), and ET by Kitamura and Stutzer (1997) and Imbens, Spady, and Johnson (1998).”
Efficient two-sided nonsimilar invariant tests in IV regression with weak instruments (2008, methodology): “. Three test statistics that are functions of Q are the LM statistic of Kleibergen (2002) and Moreira (2001) , the Anderson and Rubin (1949) statistic (AR), and the Moreira (2003) conditional likelihood ratio statistic (CLR).”
Performance of Conditional Wald Tests in IV Regression with Weak Instruments (2007, background): “The LR statistic The other test considered in this study is Moreira’s (2003) CLR test, which is based on the statistic d LR 1⁄4 1 2 fQ̂S Q̂T þ 1⁄2ðQ̂S þ Q̂TÞ2 4ðQ̂SQ̂T Q̂ 2 STÞ g.”
Weak Identification in Fuzzy Regression Discontinuity Designs (2016, unknown): “Such critical values can be generated using the approach of Moreira (2001, 2003) by conditioning on a statistic that is (i) asymptotically independent of Yn − β Xn, and (ii) summarizes the information on the strength of identification (see also Andrews, Moreira, and Stock 2006; Mills, Moreira, and…”
Credit-Induced Boom and Bust (2017, unknown): “We also verified that all our results were robust to weak instruments by employing the approach in Moreira (2009), which produces tests and confidence sets with correct size when instruments are arbitrarily weak for the just-identified case of a single endogenous variable.”
Generic results for establishing the asymptotic size of confidence sets and tests (2020, unknown): “…subscripts n are omitted.8 18 We define the Anderson and Rubin (1949) statistic and the Lagrange Multiplier statistic of Kleibergen (2002) and 19 Moreira (2009), generalized to allow for heteroskedasticity, as follows: 20 AR ≡ n̂g ′Ω̂−1̂g and LM ≡ n̂g ′Ω̂−1/2PΩ̂−1/2D̂Ω̂ −1/2̂g, where 21 D̂ ≡ Ĝ…”
Bootstrap validity for the score test when instruments may be weak (2009, result): “There is a slight difference between Moreira’s (2001, 2003) design and ours. In perfect analogy with our results, our design takes Z as being random whereas Moreira’s (2001, 2003) design takes Z as being fixed. The first-stage F-statistic corresponds to the concentration parameter λ′λ/k in the notat”
Further results on projection-based inference in IV regressions with weak, collinear or missing instruments ∗ (2007, background): “In the framework of linear simultaneous equations and in view of weak instrument problems, the importance of using pivotal functions for statistical inference has been recently reemphasized by several authors [see Dufour (1997), Staiger and Stock (1997), Wang and Zivot (1998), Zivot et al. (1998), S”