Tests and Confidence Sets with Correct Size when Instruments are Potentially Weak

By A. Mikusheva, Brian P. Poi. 2006

104 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/b5bc54ab7bbee67c6d670f74bca39ef81c2e839e

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

How Important Are Financing Constraints? The Role of Finance in the Business Environment (2008, unknown): “Following Moreira and Poi (2001) and Mikusheva and Poi (2001) we obtain”
Causal Inference with Observational Data (2007, unknown): “Perhaps best of all, you can conduct inference that is robust to weak instruments; with one endogenous variable, use condivreg (Mikusheva and Poi 2006), or with more than one, use tests described by Anderson and Rubin (1949) and Baum, Schaffer, and Stillman (2007, sections 7.4 and 8).”
Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology (2008, unknown): “All of the IV analyses were performed in Stata (version 9.2) using the user-written ivreg2 command [56] (usefully, Stata is one language that many epidemiologists and econometricians have in common).”
The Effects of Rural Electrification on Employment: New Evidence from South Africa (2011, background): “These AR confidence intervals have correct coverage properties in the presence of weak instruments while standard Wald tests do not (Mikusheva and Poi, 2006; Chernosukov and Hansen, 2007).”
Internationalization and Innovation of Firms: Evidence and Policy (2013, unknown): “Employees (20-49) 168 1,150 1,135 176 1,407 1,244 805 6,085”
Using multiple genetic variants as instrumental variables for modifiable risk factors (2012, background): “Alternative IV estimators to TSLS may have better finite sample properties when instruments explain a small proportion of phenotypic variability.(59,60)”
Enhanced routines for instrumental variables/GMM estimation and testing (2007, background): “The test provided in condivreg (Moreira and Poi (2003), Mikusheva and Poi (2006)) is uniformly most powerful in the situation where there is one endogenous regressor and i.i.d. errors.”
Enhanced Routines for Instrumental Variables/Generalized Method of Moments Estimation and Testing (2007, methodology): “The best-known approach to this problem in econometrics is that of Newey and West (1987b), which generates Ŝ by using the Bartlett kernel function and a user-specified value of q. For the Bartlett kernel, κ(·) = (1 − j/qn) if j ≤ qn − 1, 0 otherwise. These estimates are said to be heteroskedasticity”