Testing Instrument Validity for LATE Identification Based on Inequality Moment Constraints

By Martin Huber, Giovanni Mellace. Review of Economics and Statistics, 2015

128 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/46b1a96ce03d18ee74939b7be494d3d2adf8c9c6

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

Citations in the corpus

Partial Identification of Local Average Treatment Effects With an Invalid Instrument (2013, background): “As we further discuss in Section 4, the bounds on LNATEnt and LNATEat can shed light on the validity of the exclusion restriction assumption, as this assumption implies that LNATEnt = LNATEat = 0 (e.g., Hirano et al. 2000; Huber and Mellace 2010, 2011; Mealli and Pacini 2012).”
Testing exclusion restrictions and additive separability in sample selection models (2014, methodology): “A further complication is that outcomes are only observed for those with S = 1.3 Table 2 3 This issue does not arise in the endogenous treatment framework of Huber and Mellace (2011), where all outcomes are observed.”
Using Instrumental Variables for Inference About Policy Relevant Treatment Effects (2018, unknown): “2, the assumption that the index of the selection equation is additively separable as in (2) is equivalent to the assumption that Z0 affects D in the monotonic sense introduced in Imbens and Angrist (1994). Hence, I.”
External Validity in Fuzzy Regression Discontinuity Designs (2019, unknown): “See for example, Huber and Mellace (2015), Angrist et al. (2016), Battistin, De Nadai, and Vuri (2017), and Black et al. (2017).”
Evaluating Public Programs with Close Substitutes: The Case of Head Start (2016, background): “This is analogous to the refutability of IV restrictions in cases with a large reduced form but small first stage (Kitigawa, forthcoming; Huber and Mellace, 2015).”
Judging Judge Fixed Effects (2019, background): “The test does not have power against all violations of Condition 1, however, as is also the case for other specification tests in the literature (Kitagawa, 2015; Huber and Mellace, 2015; Mourifié and Wan, 2017).”