[11月22日]计量、金融和大数据分析workshop

发布日期:2019-11-18 09:41    来源:

【题目】: Panel Threshold Models with Interactive Fixed Effects

【主讲人】:李鲲鹏

教授 博士生导师 首都经济贸易大学

【时间】:2019年11月22日(周五)10:00-11:30

【地点】:北大经济学院107室

【摘要】: This paper studies estimation and inference in a panel threshold model in the presence of interactive fixed effects. We study the asymptotic properties of the least squares estimators of the regression parameters in the model in the shrinking-threshold-effect framework. We find that under some conditions the threshold parameter can be estimated at a convergence rate faster than the usual parameter rate so that its estimation has asymptotically negligible role on the estimation of the slope coefficients in the model. The inference on the threshold parameter can be conducted based on a likelihood ratio test statistic as in the cross-sectional or time series setup. We also propose a test for the presence of threshold effect. Monte Carlo simulations suggest that our estimators and test statistics perform well in finite samples. We apply our method to study the effect of financial development on economic growth and find that there is indeed a turning point in the effect for all three measures of financial development when the cross-sectional dependence is properly accounted for.

【主讲人简介】:

李鲲鹏

教授 博士生导师 国际经济管理学院 首都经济贸易大学;

经济学博士 清华大学经济管理学院 2011年;

研究领域包括高维因子模型、交互效应面板数据模型、空间计量模型、动态面板模型、时间序列模型、非参数计量模型、经验过程的理论与应用;

他目前已经在中英文期刊上发表论文20多篇。他的论文发表于《Journal of Business & Economic Statistics》,《Review of Economics and Statistics》,《Journal of Econometrics》,《Annals of Statistics》,《统计研究》等国内外优秀期刊上。