南京审计大学林金官教授学术报告 3月16日上午
发布时间: 2019-03-15 访问次数: 185

报告题目【Semiparametric Stochastic Volatility Models with Time-varying Leverage Effect: Properties and Estimation】

时间:2019年3月16日 (星期六)上午 11:00

地点:旗山校区理工北楼601报告厅

主讲:南京审计大学教授,林金官

主办:数学与信息学院

参加对象:相关教师和研究生


专家简介:林金官,男,1964年生,安徽省滁州市人,统计学博士、教授,曾任东南大学数学系教授、统计学科带头人、系副主任。现任南京审计大学统计科学与大数据研究院院长、统计与数学学院主持工作副院长,兼任中国现场统计研究会工程概率统计学会副会长、中国现场统计研究会资源与环境统计学会副会长、教育部统计学类教学指导委员会委员、中文核心期刊《应用概率统计》与《数理统计与管理》杂志编委等。主持省部级以上课题16项,其中国家自然科学基金和国家社会科学基金4项。已在东南大学培养博士生13名、硕士生数十名。


报告摘要:Asymmetric stochastic volatility models mimic many of the stylized facts attributed to financial time series of asset returns. In this paper, a new generalized time-varying asymmetric stochastic volatility (GTVASV) model is proposed, in which the distribution of the return innovation is generalized to an unknown distribution and the correlation structure between the return innovation and volatility disturbance is time-varying and nonparametrically depends on the type of news arrived to the market. Statistical properties of the proposed model are derived. The model estimation are carried out by Bayesian MCMC method. The performance of the resulting estimates is examined in simulations. The proposed model is fitted to time series of compound daily log-returns for stock index & P500 and the continuously compounded daily returns for the individual stock Microsoft (MSFT). Implementation on empirical studies also illustrates the suitability of the proposed model and the estimation method in practice.