学术报告新

Xiaojun Song is an Associate Professor in the Department of Business Statistics and Econometrics at the Guanghua School of Management, Peking University. He holds a PhD in Economics from Universidad Carlos III of Madrid, Spain. His research interests lie in theoretical econometrics, including nonparametric and semiparametric methods, hypothesis testing and bootstrapping, as well as applied econometrics.

Varying coefficient models provide a robust and practical framework while balancing flexibility and simplicity. This study discusses a series of hypothesis testing problems in varying coefficient models, including a constancy test, a significance test, and a partial constancy test. The null hypotheses are formulated as conditional moment restrictions to derive the integral conditional moment (ICM)-type test statistics. The asymptotic properties of the proposed tests are established, and critical values are computed by easy-to-implement multiplier bootstrap procedures. Finite sample properties are evaluated through Monte Carlo experiments, demonstrating appropriate size and power performance in finite samples.