Premium
Treatment Effect Estimation Using Nonlinear Two‐Stage Instrumental Variable Estimators: Another Cautionary Note
Author(s) -
Chapman Cole G.,
Brooks John M.
Publication year - 2016
Publication title -
health services research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.706
H-Index - 121
eISSN - 1475-6773
pISSN - 0017-9124
DOI - 10.1111/1475-6773.12463
Subject(s) - instrumental variable , observational study , econometrics , nonlinear system , average treatment effect , estimator , statistics , mathematics , residual , omitted variable bias , contrast (vision) , estimation , population , linear model , outcome (game theory) , variable (mathematics) , treatment effect , computer science , medicine , economics , algorithm , physics , management , environmental health , mathematical economics , quantum mechanics , artificial intelligence , mathematical analysis , traditional medicine
Objective To examine the settings of simulation evidence supporting use of nonlinear two‐stage residual inclusion (2 SRI ) instrumental variable ( IV ) methods for estimating average treatment effects ( ATE ) using observational data and investigate potential bias of 2 SRI across alternative scenarios of essential heterogeneity and uniqueness of marginal patients. Study Design Potential bias of linear and nonlinear IV methods for ATE and local average treatment effects ( LATE ) is assessed using simulation models with a binary outcome and binary endogenous treatment across settings varying by the relationship between treatment effectiveness and treatment choice. Principal Findings Results show that nonlinear 2 SRI models produce estimates of ATE and LATE that are substantially biased when the relationships between treatment and outcome for marginal patients are unique from relationships for the full population. Bias of linear IV estimates for LATE was low across all scenarios. Conclusions Researchers are increasingly opting for nonlinear 2 SRI to estimate treatment effects in models with binary and otherwise inherently nonlinear dependent variables, believing that it produces generally unbiased and consistent estimates. This research shows that positive properties of nonlinear 2 SRI rely on assumptions about the relationships between treatment effect heterogeneity and choice.