地点: 学正楼212
邀请人：统计与金融数学研究室 周秀轻教授
报告摘要： This paper introduces a model that can accommodate both continuous-time- diffusion and discrete-time GARCH model with additive jumps by embedding the discrete GARCH structure with jumps in the continuous volatility process. The key feature of the proposed model is that the corresponding conditional integrated volatility adopts the GARCH structure that takes into account the effect of jumps on future volatility. We pro- pose a Griddy-Gibbs sampler approach to estimate parameters, and then develop volatil- ity forecasting and Value at Risk (VaR) forecasting based on the Peaks Over Threshold (POT). Simulations are carried out to check the finite sample performance of the proposed methodology, and an empirical study with Microsoft Corporation Stock Price shows that the volatility is heavily influenced by the extreme reactions than the continuous innovations. We find that both the simulation and empirical example results in most cases support the proposed model.
报告人简介：林金官博士，统计学教授、博士生导师。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与数学学院院长、统计科学与大数据研究院院长。主要从事非线性统计、计量经济、金融统计与风险度量、统计诊断、面板数据分析和统计应用等方面的研究工作。本科、硕士分别毕业于华东师范大学数学与数理统计专业，博士毕业于东南大学系统工程专业。2000年以来,在国内外核心期刊上发表论文一百余篇，其中SCI和SSCI收录论文近百篇。目前担任2018-2022教育部统计学类教学指导委员会委员、全国工业统计教学研究会副会长、中国现场统计研究会工程概率统计学会副理事长、中国现场统计研究会资源与环境学会副理事长、江苏省概率统计学会秘书长、中文核心期刊《系统科学与数学》、《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18项，其中国家自然科学基金和国家社会科学基金5项。已培养博士生13名、硕士生数十名。
