Beyond First-Order Bias in Predictive Regressions: Estimation, Inference, and Return Predictability
Prof. Jun YU
UMDF Chair Professor of Finance and Economics
Chair Professor of Finance and Economics
Dean of Faculty of Business Administration
University of Macau
Date: 11 March 2026 (Wednesday)
Time: 13:00-14:00
Venue: E22-Lobby
Moderator: Prof. Shen ZHAO, Associate Professor in Finance
Abstract
This paper reveals a large and overlooked source of distortion in predictive regressions: higher-order bias. While existing corrections only address first-order bias, we show that higherorder terms in both direct regression (DR) and implication (IM) based estimators can seriously mislead both statistical and economic conclusions, especially with persistent predictors and long horizons. We then develop iBoot, a unified framework that integrates indirect inference with bootstrap methods. iBoot (i) jointly corrects first- and higher-order biases in both estimation and inference for DR and IM estimators, and (ii) applies seamlessly to both short- and long-horizon regressions. Simulations demonstrate that iBoot achieves minimal bias, accurate confidence intervals, correct test size, and high power. When applied to classic return predictors, iBoot materially changes the economic significance of return predictability. This paper thus establishes a new foundation for credible and robust estimation and inference in predictive regressions.
Speaker
Prof. Jun YU, a financial econometrician, currently serves as the Dean of the Faculty of Business Administration at the University of Macau and he is also appointed as the University of Macau Development Foundation Chair Professor of Finance and Economics. He serves as an Associate Editor for prestigious international journals such as the Journal of Econometrics and Econometric Theory. Furthermore, Prof. Jun YU is a founding member of the International Society for Financial Econometrics and a Fellow of The Journal of Econometrics. Professor Jun YU obtained his PhD degree in Economics from the University of Western Ontario in Canada. He has published over 90 articles in top international academic journals such as the Review of Financial Studies, Journal of Econometrics, etc. Collaborating with scholars from the United States and Australia, he has developed a tool to measure exuberance in financial markets, which have been widely adopted across countries and regions in the US, Europe, and Asia, serving as a valuable resource for identifying financial crises and asset bubbles, as well as informing economic policies.
All are welcome!
