A New Test for Multiple Predictive Regression, a paper co-authored by our school’s Assistant Professor GUO Junjie and Professor XU Ke-Li from Indiana University, was accepted by Journal of Financial Econometrics, a world-class journal on econometrics.

Predictive regression poses challenges to statisticians and empirical economists in finance and macroeconomics due to strong persistence of many potentially useful predictors. Strongly persistent predictors, unless strictly exogenous, affect textbook standard econometric procedures, causing biased estimation of the regression slope and invalid inference.

This paper considers inference for predictive regressions with multiple predictors. Extant tests for predictability (especially for joint predictability) may have unsatisfactorily large size distortion and tend to discover spurious predictability as the number of predictors increases. To tackle this problem, we propose a test based on the few-predictors-at-a-time parsimonious system approach.

Empirical Monte Carlos demonstrate the remarkable infinite-sample performance regardless of numerosity of predictors and their persistence properties. Empirical application to equity premium predictability is provided.