Finite Sample Properties of Some Alternative Gmm Estimators

By L. Hansen, John Heaton, Amir Yaron. 2015

627 citations recorded by the source corpus.

We investigate the small-sample properties of three alternative generalized method of moments (GMM) estimators of asset-pricing models. The estimators that we consider include ones in which the weighting matrix is iterated to convergence and ones in which the weighting matrix is changed with each choice of the parameters. Particular attention is devoted to assessing the performance of the asymptotic theory for making inferences based directly on the deterioration of GMM criterion functions.

Subfields: Control and Systems Engineering, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/8530b64dd14346a351dfe95cc2af2d992c5da24f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Statistics and Probability

Citations in the corpus

An M‐estimator of spatial tail dependence (2014, methodology): “tep procedure where the weight matrix ( ) 1 is recalculated within the minimisation routine. This resembles, but is substantially dierent from a continuously updating generalised method of moments (Hansen et al., 1996). Rather than as in equation (3.3), the pairwise M-estimator of 0 would be dened ”
Covariate balancing propensity score (2014, methodology): “surveys when some auxiliary information about the population distribution is available (see Deming and Stephan (1940), Oh and Scheuren (1983), Little and Wu (1991), Hellerstein and Imbens (1999), Nevo (2003) and many others).”
Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes (2019, methodology): “Note that in settings (1) and (3), ˆ ρ DM is eﬃcient for M 2 q per Section 4.2 (or almost so; it would be eﬃcient if we used eﬃcient GMM instead of one-step GMM).”
Inference for Iterated GMM Under Misspecification (2021, methodology): “We use (26) for our numerical and empirical calculations.”
A review of instrumental variable estimators for Mendelian randomization (2015, methodology): “This is known as the continuous updating estimator (CUE), as the weighting matrix updates during the estimation procedure.(82) The CUE is similar to the LIML estimator in terms of confidence intervals and finite sample bias,(48) and the two Burgess et al.”
Caught in the crossfire: Dimensions of vulnerability and foreign multinationals' exit from war‐afflicted countries (2017, positive): “We also conducted tests of identification, as weak instruments can distort test statistics and cause misleading statistical inference (Hansen, Heaton, and Yaron, 1996).”
A continuous updating weighted least squares estimator of tail dependence in high dimensions (2016, methodology): “The construction of the estimator bears some similarity with the continuous updating generalized method of moments (Hansen et al., 1996); the present estimator, however, is substantially diﬀerent and does not use moments.”
Estimation of linear dynamic panel data models with time‐invariant regressors (2019, neutral): “In order to overcome the weak instruments problem in the context of estimating the effects of time-varying regressors, Hsiao et al. (2002) propose a transformed likelihood approach that is based on the model in first differences.”
In Search of the Armington Elasticity (2014, neutral): “Combining the foregoing expression with the demand V jj gt as computed from equation (10), we obtain (34):”