报 告 人：刘 志 副教授（澳门大学）
报告题目：Data-Efficient Realized Covariance Matrix
报告时间：2018年6月14日（周四）10:30-11:30
报告地点：静远楼1506报告厅
主办单位：数学与统计学院、科学技术研究院
报告人简介：
刘志，澳门大学数学系副教授。主要研究方向包括: 金融超高频数据分析、随机过程统计推断，等。其研究近年来获得了澳门政府以及国家自然科学基金等多项基金的资助，在统计学、金融和生物信息国际期刊发表论文40余篇，主要研究成果发表在AoS、JASA、JoE、JBES、Bioinformatics、SPA、ET等相关研究方向的权威期刊上。
报告摘要：
The variate features of high frequency data, such as, market microstructure effect, asynchronous trading, multiple records, etc, bring challenges in the estimation of covariance matrices among assets. In this paper, we proposed a data-efficient estimator which utilizes all of data, whereas all existing approaches discard part of data, to deal with the asynchronicity and multiple records. By congregating the data points within the synchronized time intervals as average, we found that the realized covariance is still consistent. We have established the related central limit theorem and its studentized version. Compared to the existing approaches, our estimator achieves much improvement in the estimation efficient for those highly traded liquid assets. Through a variety of synthetic data experiments, we assess the finite sample performance of proposed estimator and make comparison with other existing methods. Finally, we illustrate the estimator via some real data analyses.
