Optimal Two‐Sided Invariant Similar Tests for Instrumental Variables Regression

By Donald W. K. Andrews, Marcelo J. Moreira, James H. Stock. 2006

341 citations recorded by the source corpus.

Subfields: Finance, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/76123554d78895d08f21faa6796b1fc24f9396d5

How researchers see this paper

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

In Finance

In Statistics and Probability

Citations in the corpus

Efficient two-sided nonsimilar invariant tests in IV regression with weak instruments (2008, methodology): “On the latter point, the one-sided and two-sided testing problems turn out to be quite different when the instruments are weak, and conclusions in the twosided problemdonot necessarily carry over to the one-sided case, see Andrews et al. (2007b).”
Global Trade and the Maritime Transport Revolution (2010, methodology): “…freight is now small, negative, and statistically indistinguishable from 0.4 The bottom rows of table 4 give an alternative IV estimator and test for the coefficient on freight: the Moreira (2003) conditional IV estimator and the Andrews, Moreira, and Stock (2006) conditional likelihood ratio test.”
Does economic freedom cause prosperity? An IV approach (2009, methodology): “To evaluate the robustness of our test results, we performed a conditional likelihood ratio test originally developed by Morreira (2003) and further refined by Andrews and Stock (2005) and Andrews et al. (2006) .”
Performance of Conditional Wald Tests in IV Regression with Weak Instruments (2007, methodology): “The power of the resulting conditional likelihood ratio (CLR) test has been examined in detail in Andrews et al. (2006) (hereafter, AMS), however conditional Wald (CW) tests have not been examined numerically.”
Decision Theory Applied to an Instrumental Variables Model (2007, result): “There is some overlap in the Section 8 result on hypothesis tests with independent work by Andrews, Moreira, and Stock (2006).”
Corruption is bad for growth (even in the United States) (2011, methodology): “…then estimate unbiased effects of corruption on growth and investment using both two-stage least squares (2SLS) and limited information maximum likelihood (LIML) methods along with the conditional likelihood ratio (CLR) test to control for weak instruments (Moreira 2003, Andrews et al. 2006, 2007).”
Testing with many weak instruments (2007, methodology): “(7) The AR, LM, and LR test statistics can be written asdARn ¼ bQS;n=k,dLMn ¼ bQ2ST ;n= bQT ;n, and cLRn ¼ 1 2 bQS;n bQT ;n þ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffið bQS;n bQT ;nÞ2 þ 4 bQ2ST ;nq”
Conditional Inference With a Functional Nuisance Parameter (2014, neutral): “Consequently, when applied to the QLR statistic in homoskedastic linear IV, our approach yields the CLR test of Moreira (2003), which Andrews, Moreira, and Stock (2006) shows is nearly a uniformly most powerful test in a class of invariant similar two-sided tests in the homoskedatic Gaussian linear…”
Instrumental Variables: An Econometrician's Perspective (2014, neutral): “Methodological studies considering essentially this set of assumptions, sometimes without explicitly stating instrument relevance, and sometimes adding additional assumptions, include Robins (1989), Heckman (1990), Manski (1990), Imbens and Angrist (1994), Angrist, Imbens and Rubin (1996), Robins an”
Uniformity and the Delta Method (2015, methodology): “Important examples include the study of ‘local-to-zero’ behavior of estimators, tests and confidence sets in linear instrumental variable regression (Staiger and Stock 1997; Moreira 2003; Andrews et al. 2006); the ‘local-tounit root’ analysis for autoregressive parameters (see Stock and Watson 1996;”
ESTIMATION AND INFERENCE WITH WEAK, SEMI-STRONG, AND STRONG IDENTIFICATION (2010, neutral): “The CLR test is approximately asymptotically optimal under weak and strong identification in the classes of invariant similar and invariant tests; see Andrews, Moreira, and Stock (2006, 2008) and Chernozhukov, Hansen, and Jansson (2009).”
Exporting and Firm Performance: Chinese Exporters and the Asian Financial Crisis (2010, neutral): “Due to concerns that the set of instruments may be relatively weak, the bottom row of table 7 presents regression results where standard errors and 5% significance tests follow the methodology to calculate conditional IV estimates proposed by Andrews et al. (2006). Their method adjusts critical valu”
The Hausman Test and Weak Instruments (2011, background): “7 (b) under the alternative hypothesis ( 3 ),”
Generic results for establishing the asymptotic size of confidence sets and tests (2020, neutral): “10 More extensive tables of critical values are given in the Supplemental Material to Andrews et al. (2006), which is available on the Econometric Society website.”
Weak Identification in Fuzzy Regression Discontinuity Designs (2016, neutral): “…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 Vilela 2014).”
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).”