SGMM: Stochastic Approximation to Generalized Method of Moments
Authors:
Preprint
Abstract
We introduce a new class of algorithms, Stochastic Generalized Method of Moments (SGMM), for estimation and inference on (overidentified) moment restriction models. Our SGMM is a novel stochastic approximation alternative to the popular Hansen (1982) (offline) GMM, and offers fast and scalable implementation with the ability to handle streaming datasets in real time. We establish the almost sure convergence, and the (functional) central limit theorem for the inefficient online 2SLS and the efficient SGMM. Moreover, we propose online versions of the Durbin-Wu-Hausman and Sargan-Hansen tests that can be seamlessly integrated within the SGMM framework. Extensive Monte Carlo simulations show that as the sample size increases, the SGMM matches the standard (offline) GMM in terms of estimation accuracy and gains over computational efficiency, indicating its practical value for both large-scale and online datasets. We demonstrate the efficacy of our approach by a proof of concept using two well known empirical examples with large sample sizes.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Fast inference for quantile regression with tens of millions of observationsJournal of Econometrics, 2024
- Online Statistical Inference for Stochastic Optimization via Kiefer-Wolfowitz MethodsJournal of the American Statistical Association, 2024
- Online Covariance Matrix Estimation in Stochastic Gradient DescentJournal of the American Statistical Association, 2021
- HAR Inference: Recommendations for PracticeJournal of Business & Economic Statistics, 2018
- Sieve inference on possibly misspecified semi-nonparametric time series modelsJournal of Econometrics, 2014
- EDGEWORTH EXPANSIONS FOR SPECTRAL DENSITY ESTIMATES AND STUDENTIZED SAMPLE MEANEconometric Theory, 2001
- Simple Robust Testing of Regression HypothesesEconometrica, 2000
- Nonparametric Adaptive Learning with FeedbackJournal of Economic Theory, 1998
- Artificial neural networks: an econometric perspectiveEconometric Reviews, 1994
- Adaptive Algorithms and Stochastic Approximations1990