This document provides an overview of enhanced routines for instrumental variables (IV) and generalized method of moments (GMM) estimation and testing in Stata. The authors, Christopher F. Baum, Mark E. Schaffer, and Steven Stillman, extend their 2003 paper on IV and GMM estimation, presenting new features and improvements to the ivreg2 suite of commands. These enhancements include robust estimation and testing that account for heteroskedasticity and autocorrelation, as well as options for weak instrument handling, limited-information maximum likelihood, and k-class estimation. The paper also discusses the Frisch–Waugh–Lovell (FWL) theorem and its implementation in ivreg2, which allows for partialling out exogenous regressors to address rank-deficient covariance matrices. The authors provide detailed explanations of the theoretical foundations and practical applications of these enhancements, along with examples using real datasets to illustrate the commands and their usage.This document provides an overview of enhanced routines for instrumental variables (IV) and generalized method of moments (GMM) estimation and testing in Stata. The authors, Christopher F. Baum, Mark E. Schaffer, and Steven Stillman, extend their 2003 paper on IV and GMM estimation, presenting new features and improvements to the ivreg2 suite of commands. These enhancements include robust estimation and testing that account for heteroskedasticity and autocorrelation, as well as options for weak instrument handling, limited-information maximum likelihood, and k-class estimation. The paper also discusses the Frisch–Waugh–Lovell (FWL) theorem and its implementation in ivreg2, which allows for partialling out exogenous regressors to address rank-deficient covariance matrices. The authors provide detailed explanations of the theoretical foundations and practical applications of these enhancements, along with examples using real datasets to illustrate the commands and their usage.