报告内容：Heterogeneity and endogeneity become increasingly common in econometric practice. Threshold regression provides a simple yet flexible modeling strategy to account for heterogeneity by allowing for increased threshold effects in functional form. This paper studies estimation and inference for multiple threshold regression with endogenous regressors. We exploit a novel two-stage estimation procedure based on instrumental variables, which first identifies a set of possible threshold locations using the group LASSO estimation, and then refines the candidate set by a predetermined selection criterion. Given that the performance of conventional information criterion is sensitive to the choice of penalization factor, we develop a data-adaptive threshold-based cross-validation criterion incorporating an order-preserved sample-splitting strategy to determine the number of thresholds. In cases where regressors and threshold variable are both endogenous, the proposed approaches remain applicable with slight adjustments using the control function framework. Numerical examples demonstrate the favourable performance of our proposals, including an application to the threshold effect of 401(k) plans on the wealth.
报告人：万闯
邀请人：孙军
时间：2025年3月26日（周三）14:00-17:00
地点：西校图书馆三楼研讨室2（线下）；腾讯会议（835-803-837）
主办：统计与应用数学学院、科研处
报告人简介：万闯，暨南大学经济学院统计与数据科学系助理教授。2022年7月毕业于厦门大学邹至庄经济研究院，2022年7月到2024年7月在南开大学统计与数据科学学院开展博士后研究。在Journal of Econometrics、Journal of Business and Economic Statistics、Biometrics、Scandinavian Journal of Statistics等期刊发表论文十余篇；主持国家自然科学青年基金一项以及博士后面上项目两项；主要研究方向为可信赖的数据科学理论与方法及其在计量经济学中的应用。