Addressing Endogeneity Using a Two-Stage Copula Generated Regressor Approach

作者：Fan Yang, Yi Qian, Hui Xie · 发表于：Journal of Marketing Research · 年份：2024 · DOI：10.1177/00222437241296453 · 被引用次数：25 · 研究领域：Neural Networks and Applications

The ubiquitous presence of endogenous regressors presents a significant challenge when drawing causal inferences using observational data. The classical econometric method used to handle regressor endogeneity requires instrumental variables (IVs) that must satisfy the stringent condition of exclusion restriction, rendering it unfeasible in many settings. Herein, the authors propose a new IV-free method that uses copulas to address the endogeneity problem. Existing copula correction methods require nonnormal endogenous regressors: Normally or nearly normally distributed endogenous regressors cause model nonidentification or significant finite-sample bias. Furthermore, existing copula control function methods presume the independence of exogenous regressors and endogenous regressors. The authors' generalized two-stage copula endogeneity-correction (2sCOPE) method simultaneously relaxes the two key identification requirements while maintaining the Gaussian copula regressor-error dependence structure. They prove that under the Gaussian copula dependence structure, 2sCOPE yields consistent causal-effect estimates with correlated endogenous and exogenous regressors as well as normally distributed endogenous regressors. In addition to relaxing the identification requirements, 2sCOPE has superior finite-sample performance and addresses the significant finite-sample bias problem due to insufficient regressor nonnormality. Moreover, 2sCOPE employs generated regressors derived from exis...