报告时间：2026年11月2日（周一）下午 15:00-16:00

报告地点：苏州大学天赐庄校区精正楼306

报告人：宋晓军 副教授，北京大学

报告摘要：

Many structural and dynamic economic models imply that key parameters are identified by conditional quantile restrictions. Building on the exponential-weighting approach of \citet{bierens1990} and the penalized maximum statistics of \citet{chen2025}, we develop uniform inference for such parameters. We construct an adaptive $\ell_1$-penalized supremum statistic from a process indexed jointly by quantiles and weighting directions. The statistic aggregates evidence across quantiles and regularizes maximization over directions. A shifted bootstrap estimates local power for penalty selection. Under the stated conditions, adaptive selection asymptotically preserves or improves worst-case local power relative to the unpenalized test over a prespecified set of alternatives. The improvement is strict whenever a positive candidate penalty raises the population maximin criterion above its value at zero. We extend the process and bootstrap theory to settings with pre-estimated nuisance parameters, characterize their contribution to the limiting process, correct the variance estimator, and establish Gaussian multiplier bootstrap validity for null inference. Monte Carlo comparisons favor CvM--KS for size control under pre-estimation and show that adaptive penalization can improve empirical power relative to the unpenalized test.

报告人简介：

宋晓军，北京大学光华管理学院商务统计与经济计量系副教授，博士生导师，西班牙马德里卡洛斯三世大学经济学博士。主要研究兴趣是理论计量经济学，包括非参数/半参数方法，假设检验和自助法，以及计量经济学的应用等。论文发表在Econometric Reviews, Econometric Theory, Journal of Applied Econometrics, Journal of Business & Economic Statistics, Journal of Econometrics和Management Science等国际期刊。主持和参加自然科学基金面上项目和重点项目等。获得北京大学优秀班主任、北京大学优秀博士学位论文指导教师、北京大学蔡元培美育奖教金等荣誉。自2020年1月起，担任Economic Modelling副主编。

邀请人：刘洋