报告人：张园园，清华大学
时间：2020.11.26 20:00-21:00
报告摘要：
A time varying ARCH model isproposed to describe the changing volatility of a financial return seriesover long time horizon, along with two-step least squares and maximumlikelihood estimation procedures. After preliminary estimation of the timevarying trend in volatility scale, approximations to the latent stationaryARCH series are obtained, which are used to compute the least squaresestimator (LSE) and maximum likelihood estimator (MLE) of the ARCHcoefficients. Under elementary and mild assumptions, oracle efficiency ofthe two-step LSE for ARCH coefficients is established, i.e., the two-stepLSE is asymptotically as efficient as the infeasible LSE based on theunobserved ARCH series. As a matter of fact, the two-step LSE deviates fromthe infeasible LSE by opn-12. The two-step MLE,however, does not enjoy such efficiency, but n1/2 asymptotic normalityis established for both the two-step MLE as well as its deviation from the infeasible MLE. Simulation studies corroborate the asymptotic theory, andapplication to the S&P 500 index daily returns from 1950 to 2018 indicatessignificant change in volatility scale over time.
报告人简介：
张园园，清华大学统计学研究中心博士，受国家留学基金委资助赴爱荷华州立大学联合培养一年；主要研究方向为非参数与半参数统计推断、时间序列分析以及函数型数据分析等。学术论文在TEST、Journal of Time Series Analysis和Journal of Neuroscience Methods等学术期刊发表
欢迎参加！
邀请人：马学俊