Title: Uniform Inference for Parameters Identified by Conditional Quantile Restrictions

Speaker: Song Xiaojun, Associate Professor and Doctoral Supervisor, Department of Business Statistics and Econometrics, Guanghua School of Management, Peking University. He holds a doctoral degree in Economics from Universidad Carlos III de Madrid, Spain.

Abstract: This paper develops a new inference procedure for parameters identified by conditional quantile restrictions. Building on Bierens (1990) and Chen et al. (2025), we introduce a quantile-based penalized Bierens maximum statistic and propose a multiplier bootstrap method for valid and powerful uniform inference on the parameter function. The proposed test is direct and independent of any preliminary parameter estimation step, ensuring computational simplicity and robustness. Furthermore, we provide a data-dependent rule for selecting the optimal penalty parameter by choosing the penalty that maximizes local power under the sequence of local alternatives. Monte Carlo simulations based on quantile autoregressive models demonstrate that the procedure not only maintains correct size but also achieves higher power compared with existing methods.

Date & time: 24 April 2026, 15:00 - 16:30

Venue: B321, Zhixin Building, Central Campus, Shandong University