Empirical likelihood estimation and consistent tests with conditional moment restrictions

By Stephen G. Donald, Guido W. Imbens, Whitney K. Newey. 2003

221 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/672bd71fc101a2566d135e18b70e439cfc76e3f4

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

GMM with many weak moment conditions (2005, result): “The following result improves upon Theorem 6.2 of Donald, Imbens, and Newey (2003) .”
A regularization approach to the many instruments problem (2012, methodology): “Donald and Neweys and our three estimators are asymptotically equivalent since they all attain the semiparametric e¢ ciency bound. We compare their small sample properties using both the theoretical expression of the MSE and Monte Carlo experiments. The theoretical MSE shows that the Tikhonov regul”
GMM with Many Moment Conditions (2004, result): “In this regard the results of the present paper differ from those of Donald, Imbens and Newey (2003) and Koenker and Machado (1999), who consider GMM and empirical likelihood estimation under strong moment conditions and explore conditions on q that permit usual asymptotic theory and statistical t”
Leave‐Out Estimation of Variance Components (2018, unknown): “The existing literature allows for either heteroscedastic errors and moderately few regressors (Donald et al., 2003, k3/n → 0) or homoscedastic errors and many regressors (Anatolyev, 2012, k/n ≤ c 1).”
Estimation of Average Treatment Effects With Misclassification (2007, methodology): “, J , we may estimate β0 using Hansen’s (1982) GeneralizedMethod ofMoments (GMM). Asymptotic efficiency can be obtained by using estimated optimal η j(Z) functions as in Newey (1993), Donald, Imbens, and Newey (2003), or Dominguez and Lobato (2004).”
Empirical Likelihood Methods in Econometrics: Theory and Practice (2006, methodology): “Donald, Imbens, and Newey (2003) apply unconditional versions of empirical likelihood and GEL (hence GMC) to the implied unconditional moment condition model.”
SMOOTHED EMPIRICAL LIKELIHOOD METHODS FOR QUANTILE REGRESSION MODELS (2006, unknown): “The SEL confidence regions considered are Ihc, I hc, and I b hc which are defined in (16), (32), and (33) respectively.”
Choosing instrumental variables in conditional moment restriction models (2009, methodology): “…28–36 p lim ( (Γ̃ ′W0Γ̃ )−1Γ̃ ′W0Υ̃W0Γ̃ (Γ̃ ′W0Γ̃ )−1 ) , Υ̃ = n∑ i=1 q̃(xi)q̃(xi)′σ 2i /n, and if the preliminary estimator uses optimal weighting we can show that this coincides with p limΩ∗ provided that K̃ increase with n in a way that the assumptions of Donald et al. (2003) are satisfied.”
SPECIFICATION TESTING IN MODELS WITH MANY INSTRUMENTS (2010, unknown): “Then, the asymptotic size of the conventional and Donald et al. (2003) J tests equals ( −1 (α)√ 1−λ ) .”
A local generalized method of moments estimator (2007, methodology): “vector valued function g. Asymptotically efficient estimates of θ given conditional moments E[g(θ,W) | Z = z)] = 0 can be obtained using variants of GMM or empirical likelihood such as Newey (1993), Donald, Imbens, and Newey (2003) , Otsu (2003), or Dominguez and Lobato (2004).”
Breaking the curse of dimensionality in nonparametric testing (2008, background): “When testing for additivity, U(θ) = Y − ∑q l=1 ml(Xl), where the unknown univariate functions ml(·) are properly normalized, see Gozalo and Linton (2001). When testing for a single-index model, U(θ) = Y −m(X ′β), for an unknown β and an unknown univariate function m(·), see Fan and Li (1996), Stute ”
Measuring Substitution Patterns in Diﬀerentiated-Products Industries (2019, unknown): “This problem has been solved in the optimal IV literature following Newey (1990), and Donald et al. (2003). In particular if the collection of L instruments Aj(xt) can arbitrarily well approximate any function a (xt), i.”