報告人：劉志 教授
報告題目：Realized Copula of Volatility
報告時間：2025年12月26日（周五）16:00-17:00
報告地點：云龍校區6號樓304報告廳
主辦單位：數學與統計學院、數學研究院、科學技術研究院
報告人簡介：
劉志, 澳門大學數學系教授。研究興趣包括隨機過程統計，金融統計，機器學習在生物信息和醫學數據方面的應用。論文發表于統計學及其交叉學科國際期刊，如統計學期刊Annals of Statistics, Journal of American Statistical Association, Journal of Business and Economic Statistics,計量經濟學期刊Quantitative Economics, Journal of Econometrics, Econometric Theory, Econometrics Journal, 金融學期刊Journal of Banking and Finance, Finance and Stochastics, 生物信息學期刊Bioinformatics, 以及機器學習會議AAAI等。主持完成國家自然科學基金和澳門政府基金10余項。
報告摘要：
Asset empirical volatility is random. The statistical inference on the volatility itself, conditional on the volatility path, has been well established. However, the inference on the distribution of the stochastic volatility is yet to be well studied, particularly for the multivariate case. In the work, we study the copula of the distributions of bivariate stochastic volatility. We propose realized copula based on estimators of the spot volatility, and establish both infill and long-span asymptotics. Simulation studies based on two well-known stocahstic volatility models, Heston model and OU process, demonstate good performance of our realized copula in capturing the joint distributional pattern of the bivariate stochastic volatility. We also implement our realized copula to a real high-frequency data including two futures contracts, the E-mini S&P 500 (ES) and the 10-Year Treasury Notes (TY). The results suggest a Gumbel copula with a parameter around 1.6 for two stochastic volatilities. This is a joint work with Kim Christensen, Wenjing Liu and Yoann Potiron.