Improved JIVE Estimators for Overidentified Linear Models with and without Heteroskedasticity

By Daniel A. Ackerberg, P. Devereux. Review of Economics and Statistics, 2009

56 citations recorded by the source corpus.

We introduce two simple new variants of the jackknife instrumental variables (JIVE) estimator for overidentified linear models and show that they are superior to the existing JIVE estimator, significantly improving on its small-sample-bias properties. We also compare our new estimators to existing Nagar (1959) type estimators. We show that, in models with heteroskedasticity, our estimators have superior properties to both the Nagar estimator and the related B2SLS estimator suggested in Donald and Newey (2001). These theoretical results are verified in a set of Monte Carlo experiments and then applied to estimating the returns to schooling using actual data.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/6a143aa12f290e57ee686b2de2957258931d8a36

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

Earnings Returns to the British Education Expansion. (2011, methodology): “Our baseline 2SLS specification uses dummy variables for being in a cohort born in 1970, 71, 72, 73, 17 Ackerberg and Devereux (2009) show that LIML can be sensitive to heteroskedasticity and suggest alternative JIVE-type estimators.”
ESTIMATION IN AN INSTRUMENTAL VARIABLES MODEL WITH TREATMENT EFFECT HETEROGENEITY (2012, background): “Therefore, the single constructed instrument is also correlated with the outcome, causing jive to be biased when the number of covariates is large (Ackerberg and Devereux, 2009).”
Evaluating Public Programs with Close Substitutes: The Case of Head Start (2016, unknown): “Offered mean Non-offered mean Differential Head Start Other centers No preschool Variable (1) (2) (3) (4) (5) (6) Male 0.”
ASYMPTOTIC DISTRIBUTION OF JIVE IN A HETEROSKEDASTIC IV REGRESSION WITH MANY INSTRUMENTS (2011, unknown): “A number of different versions of these estimators, referred to as jackknife instrumental variables (JIV ) estimators, have been proposed and analyzed by Phillips and Hale (1977), Angrist, Imbens, and Krueger (1999), Blomquist and Dahlberg (1999), Ackerberg and Devereux (2009), Davidson and MacKinno”
Controlling for endogeneity with instrumental variables in strategic management research (2008, methodology): “Finally, several authors (Ackerberg and Devereux, 2003; Chao and Swanson, 2004; Hausman et al., 2007) find that LIML estimation is sensitive to the presence of heteroscedasticity but, as Ackerberg and Devereux (2006) later noted, it is possible to derive a LIML estimator allowing for…”