Inference for Parameters Defined by Moment Inequalities Using Generalized Moment Selection

By Donald W. K. Andrews, Gustavo Soares. 2007

544 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/dfdeef58f44ebc524c3f42dd4e43cfb8295eaa55

How researchers see this paper

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

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

More on Confidence Intervals for Partially Identified Parameters (2008, methodology): “In contrast, CI(1) α does not become valid upon replacing b ∆ with ∆∗ in (3).”
Sharp identification regions in models with convex moment predictions (2010, background): “We now consider the case that players have incomplete information (see, e.g. Aradillas-López (2010), Brock and Durlauf (2001, 2007), Seim (2006), Sweeting (2009)).”
Improving the Power of Tests of Stochastic Dominance (2016, methodology): “This result is similar to (a) of Theorem 3 of Andrews and Soares (2010).”
Dynamic treatment regimes: technical challenges and applications (2014, methodology): “This partitioning is based on a “pretest” (see Olshen, 1973; Andrews, 2001b; Andrews and Soares, 2007; Cheng, 2008; Andrews and Guggenberger, 2009).”
Instrumental variables and the sign of the average treatment effect (2019, unknown): “It is possible to construct critical values that are not “least favorable” by modifying other approaches in the moment inequality literature, such as the “generalized moment selection” approach of Andrews and Soares (2010) or the recent approach by Romano et al. (2012). Indeed, “generalized moment s”
Bonferroni-Based Size-Correction for Nonstandard Testing Problems (2017, unknown): “The SLLD assumptions technically preclude certain testing problems with parameter- discontinuous null limit distributions such as testing in moment inequality models (e.g., An- drews and Soares, 2010) and certain tests allowing for weak identification (e.g., Staiger and Stock, 1997).”
Testing Instrument Validity for LATE Identification Based on Inequality Moment Constraints (2015, methodology): “Census Public Use Micro Samples (PUMS), Angrist and Evans (1998) assess the effect of fertility on labor market participation (measured by a worked for pay dummy, weeks worked, and hours per week worked) using the sex ratio of the first two children as instrument for fertility. I.e., if parents have”
Bootstrap Inference in Partially Identified Models Defined by Moment Inequalities: Coverage of the Identified Set (2010, methodology): “…our results imply that inference 8In independent research, a similar asymptotic approximation was also proposed by Soares (2006), CHT, and Andrews and Soares (2010). based on our bootstrap and our asymptotic approximation should eventually be more precise than inference based on…”
Statistical Analysis of List Experiments (2012, methodology): “10 93 /p an /m pr 04 8 proposed test, we employ the generalized moment selection (GMS) procedure proposed byAndrews and Soares(2010).”
Trucks without Bailouts: Equilibrium Product Characteristics for Commercial Vehicles (2018, methodology): “I follow Andrews and Soares (2010) in constructing these sets but with shifted rather than selected moments.”
PARTIAL IDENTIFICATION OF THE DISTRIBUTION OF TREATMENT EFFECTS AND ITS CONFIDENCE SETS (2009, background): “…Hong, and Tamer (2007), Galichon and Henry (2006), Horowitz and Manski (2000), Romano and Shaikh (2008), Stoye (2008), Rosen (2008), Soares (2006), Beresteanu and Molinari (2006), Andrews and Guggenberger (2007), Andrews and Soares (2007), Fan and Park (2007b), and Moon and Schorfheide (2007).”
Asymptotically exact inference in conditional moment inequality models (2011, background): “Arguing as in Section 6 of Armstrong (2011), this would lead to an increase in the rate at which local alternatives can approach the identified set and still be detected by a factor of log n. In contrast, the polynomial difference in the rates of convergence of the KS statistics based on integrated ”
VALIDITY OF SUBSAMPLING AND “PLUG-IN ASYMPTOTIC” INFERENCE FOR PARAMETERS DEFINED BY MOMENT INEQUALITIES (2009, methodology): “…explicit expressions are given for the probability limit of the subsampling critical value and the degree of asymptotic nonsimilarity of the subsampling test, and (v) the approach is conducive to the establishment of asymptotic local power results; as has been done in Andrews and Soares (2007).”
The Welfare Effects of Bundling in Multichannel Television Markets (2012, unknown): “Andrews and Soares (2010) introduce an alternative procedure to that in PPHI for calculating confidence sets and test statistics that are not asymptotically conservative (and, more generally, have the correct asymptotic size).”