Webbment effects (Hahn, 1998; Hirano, Imbens, & Ridder, 2003) and in the labor and industrial organization (10) literatures to estimate rich, often dynamic, structural models (Hotz and Miller, 1993; Hotz et al., 1994; Olley & Pakes, 1996; Aguir ... 4 Imbens and Wooldridge (2005) conjectured an equivalence in propensity score estimation. Webb(2003).WhenQ(x) is known or parametrically specified, however, Hahn (1998) and Hirano, Imbens, and Ridder (2003) showed that efficient estimators based on inverse propensity score weighting typically require combining a nonparametric estimate of Qˆ(x)with the known or parametrically estimated Q(x). A detailed comparison between
A Unified Framework for Efficient Estimation of General ... - arXiv
Webb(Hirano, Imbens, and Ridder,2003, Robins, Rotnitzky, and Zhao,1994, Lunceford and Davidian,2004), they show that the variance for Inverse Probability Weighting (IPW) estimators can be substantially decreased when the propensity score is estimated nonparametrically. Since their second stage also relies WebbLastly, this paper finds that adapting the efficient estimators of Hirano, Imbens, and Ridder (2003) to treatment-based sampling does not always lead to an efficient … craze slots cash fortunes
weighted least square and Difference in Difference matching
Webb236 A. ABADIE AND G. W. IMBENS Ichimura, and Todd (1998) focus on the average effect on the treated and consider estimators based on local linear kernel regression methods. Hirano, Imbens, and Ridder (2003) propose an estimator that weights the units by the inverse of their assignment probabilities and show that nonparametric se- WebbFirst, a suitable reweighting estimator is asymptotically e cient (Hirano, Imbens and Ridder 2003). Second, pair matching is asymptotically ine cient (Abadie and Imbens 2006). 2 In a recent article in the Review of Economics and Statistics, Fr olich (2004) extends the large sample WebbLinton (2002), and Hirano, Imbens, and Ridder (2003). The finite-sample properties of these estimators, however, have not been subject to extensive investigation. For their prac-tical use, it would be important to know which estimator performs best under which conditions. In addition, many of these estimators require the choice of bandwidth parame- crazer phones