SLIM: Stochastic Learning and Inference in Overidentified Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Xiaohong, Kim, Min Seong, Lee, Sokbae, Seo, Myung Hwan, Song, Myunghyun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909879452565504
author Chen, Xiaohong
Kim, Min Seong
Lee, Sokbae
Seo, Myung Hwan
Song, Myunghyun
author_facet Chen, Xiaohong
Kim, Min Seong
Lee, Sokbae
Seo, Myung Hwan
Song, Myunghyun
contents We propose SLIM (Stochastic Learning and Inference in overidentified Models), a scalable stochastic approximation framework for nonlinear GMM. SLIM forms iterative updates from independent mini-batches of moments and their derivatives, producing unbiased directions that ensure almost-sure convergence. It requires neither a consistent initial estimator nor global convexity and accommodates both fixed-sample and random-sampling asymptotics. We further develop an optional second-order refinement achieving full-sample GMM efficiency and inference procedures based on random scaling and plug-in methods, including plug-in, debiased plug-in, and online versions of the Sargan--Hansen $J$-test tailored to stochastic learning. In Monte Carlo experiments based on a nonlinear demand system with 576 moment conditions, 380 parameters, and $n = 10^5$, SLIM solves the model in under 1.4 hours, whereas full-sample GMM in Stata on a powerful laptop converges only after 18 hours. The debiased plug-in $J$-test delivers satisfactory finite-sample inference, and SLIM scales smoothly to $n = 10^6$.
format Preprint
id arxiv_https___arxiv_org_abs_2510_20996
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SLIM: Stochastic Learning and Inference in Overidentified Models
Chen, Xiaohong
Kim, Min Seong
Lee, Sokbae
Seo, Myung Hwan
Song, Myunghyun
Econometrics
Computation
Machine Learning
We propose SLIM (Stochastic Learning and Inference in overidentified Models), a scalable stochastic approximation framework for nonlinear GMM. SLIM forms iterative updates from independent mini-batches of moments and their derivatives, producing unbiased directions that ensure almost-sure convergence. It requires neither a consistent initial estimator nor global convexity and accommodates both fixed-sample and random-sampling asymptotics. We further develop an optional second-order refinement achieving full-sample GMM efficiency and inference procedures based on random scaling and plug-in methods, including plug-in, debiased plug-in, and online versions of the Sargan--Hansen $J$-test tailored to stochastic learning. In Monte Carlo experiments based on a nonlinear demand system with 576 moment conditions, 380 parameters, and $n = 10^5$, SLIM solves the model in under 1.4 hours, whereas full-sample GMM in Stata on a powerful laptop converges only after 18 hours. The debiased plug-in $J$-test delivers satisfactory finite-sample inference, and SLIM scales smoothly to $n = 10^6$.
title SLIM: Stochastic Learning and Inference in Overidentified Models
topic Econometrics
Computation
Machine Learning
url https://arxiv.org/abs/2510.20996