报告题目：Toward a uniform asymptotics for mildly explosive autoregression
报 告 人：Qiying Wang，悉尼大学
报告摘要：Mildly explosive autoregression has been extensively used in modelling the phenomenon of asset market bubbles. This paper provides a general asymptotic theory for mildly explosive autoregression. The Cauchy limit theory is confirmed to be invariant for a wide class of error processes including general linear processes with martingale difference innovations, stationary causal processes, near-epoch dependence sequence and nonlinear autoregressive time series (e.g., TAR models and Bilinear models). Our results allow the Cauchy limit theory for long memory, short memory and anti-persistent innovations in a unified framework. Extension to models with a varying drift has also been considered, which is of interest in their own right.
报告人简介：王启应，澳大利亚悉尼大学伟德国际1946源自英国教授，2007-2012 澳大利亚国家研究学者（Australian Research Fellow），2017年计量经济理论杂志(Econometric Theory） Plura Scripsit 奖的三位获奖者之一，2004年以来持续获得澳大利亚科学研究委员会(ARC)资助。王启应教授常年致力于计量经济，非参数统计，自正则化极限理论等领域的研究并做出了重要贡献，在Econometrica, Annals of Probability, Annals of Statistics, Journal of Econometrics和Econometric Theory等概率、统计和计量经济领域的顶尖期刊上发表论文80余篇，其专著《Limit theorems for nonlinear cointegrating regression》系统地介绍了非线性协整回归分析的理论体系。
报告时间：2023年10月26日 14:30-15:30
报告地点：文渊楼B408
主办单位：伟德国际1946源自英国
