Higher Order Properties of Gmm and Generalized Empirical Likelihood Estimators

By Whitney K. Newey, Richard J. Smith. 2003

943 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

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

Approximate residual balancing: debiased inference of average treatment effects in high dimensions (2016, methodology): “|X admits a well-specified logistic model the methods of Graham et al. (2012, 2016) and Hainmueller (2012) are doubly robust (Zhao and Percival, 2017); see also Hirano et al. (2001), Imbens et al. (1998) and Newey and Smith (2004).”
High dimensional generalized empirical likelihood for moment restrictions with dependent data (2013, methodology): “…of Economics, Yale University, New Haven, CT, 06520, U.S.A. Hansen and Singleton (1982) for the GMM, Owen (1988), Qin and Lawless (1994) and Kitamura (1997) for the empirical likelihood (EL), Smith (1997), Newey and Smith (2004) and Anatolyev (2005) for the generalized empirical likelihood (GEL).”
Inverse Probability Tilting for Moment Condition Models with Missing Data (2012, neutral): “Computation ofˆγ GEL would involve solving a saddle point problem with 2(K + L r ) + L t * parameters (Newey and Smith, 2004, Section 3).”
VALIDITY OF SUBSAMPLING AND “PLUG-IN ASYMPTOTIC” INFERENCE FOR PARAMETERS DEFINED BY MOMENT INEQUALITIES (2009, methodology): “19) uses a similar approach to that in Newey and Smith (2004). The proof uses the following four Lemmas.”
Empirical likelihood estimation and consistent tests with conditional moment restrictions (2003, methodology): “As shown by Newey and Smith (2002), higher order asymptotic bias of these estimators is smaller than GMM with many moments, which should lead to better asymptotic approximations to 8nite sample distributions, and bias-corrected EL is higher-order e0cient among bias-corrected estimators.”
Instrumental variables estimation with many weak instruments using regularized JIVE (2014, methodology): “CHRISTIAN HANSEN AND DAMIAN KOZBUR parameters using instrumental variables, which include IV and 2SLS, may have substantial bias when the number of instruments is not small relative to the sample size; see Bekker (1994) and Newey and Smith (2004).”
GENERALIZED EMPIRICAL LIKELIHOOD ESTIMATORS AND TESTS UNDER PARTIAL, WEAK, AND STRONG IDENTIFICATION (2003, methodology): “The main purpose then of this paper is to ascertain the performance of generalized empirical likelihood (GEL) methods [ Newey and Smith (2003) , henceforth NS] for time series instrumental variable models specified by nonlinear moment restrictions when identification may be weak [as in Stock and Wri”
Generalized Method of Moments, Efficient Bootstrapping, and Improved Inference (2002, methodology): “‚0 and by a standard mean value expan- sion, Og4 O‚5 D 4I C GH5 Og4‚05 C op41= p n5 D Op41= p n51 (A.1) so by lemma A2 of Newey and Smith (2001), it follows that O‹ exists with probability approaching 1 and O‹ DOp41= p n5.”
Can Rare Events Explain the Equity Premium Puzzle (2008, methodology): “Parker and Julliard (2005) show that testing the C-CAPM focusing on the short run correlation of consumption and returns might miss important dynamics due to slow consumption adjustment to market innovations.”
SMOOTHED EMPIRICAL LIKELIHOOD METHODS FOR QUANTILE REGRESSION MODELS (2006, positive): “Theorem 2 is a nonparametric version of Wilks' theorem (1938), which has first been proved by Owen (1991) in the standard linear regression models.”
Estimation with Weak Instruments: Accuracy of Higher-Order Bias and MSE Approximations (2004, neutral): “Combining (3), (10), (15), (17), (18), and (19), we obtain”
Generalized least squares inference in panel and multilevel models with serial correlation and fixed effects (2007, background): “Following Newey and Smith (2001), I focus on deriving the higher-order bias and variance of the estimator.”
Selecting Instrumental Variables in a Data Rich Environment (2009, methodology): “Instrumental variable selection using a ridge penalty has been suggested by Okui (2004) to solve the ‘many instrument variable’ problem. The ridge estimator differs from LASSO only in that the former replaces the L1 by an L2 penalty. Because of the nature of L2 penalty, the coefficients can only be ”