學(xué)術(shù)講座：A Factor-Based Estimation of Integrated Covariance Matrix with Noisy High-Frequency Data

金融學(xué)院SBF論壇2019年第4講

 

講座題目：A Factor-Based Estimation of Integrated Covariance Matrix with Noisy High-Frequency Data

時間：2019年3月14日（周四）12：10-13:10

地點：博學(xué)樓925

主講人： 孫宇澄

主講人簡介：

孫宇澄博士，現(xiàn)任首都經(jīng)濟(jì)貿(mào)易大學(xué)國際經(jīng)濟(jì)管理學(xué)院助理教授。他在西班牙巴塞羅那經(jīng)濟(jì)學(xué)研究生院（Barcelona Graduate School of Economics）和龐培法布拉大學(xué)（Universitat Pompeu Fabra）分別獲得金融學(xué)碩士和博士學(xué)位。他的研究興趣主要包括金融計量，高頻金融數(shù)據(jù)，非參數(shù)統(tǒng)計等。其研究成果發(fā)表在Journal of Applied Econometrics等國際知名雜志。

講座內(nèi)容簡介：

This paper studies a high-dimensional factor model with sparse idiosyncratic covariance matrix in continuous time, using asynchronous high-frequency financial data contaminated by microstructure noise. We focus on consistent estimation of the number of common factors, the integrated covariance matrix and its inverse, based on the flat-top realized kernels introduced by Varneskov (2016). Simulation results show that our estimators have good performance in finite samples. We apply our methodology to the high-frequency data on 300 liquid stocks traded in Shanghai and Shenzhen stock exchanges, and find that the model effectively captures the dynamics of volatility in the Chinese stock market.

       本文利用由于微觀結(jié)構(gòu)噪聲造成的異步高頻金融數(shù)據(jù)，研究了一個包含異質(zhì)性稀疏協(xié)方差矩陣的連續(xù)時間高緯因子模型。基于Varneskov (2016)所提出的平頂已實現(xiàn)核估計方法，本文重點對共同因子的個數(shù)，積分協(xié)方差矩陣及其逆矩陣的一致性估計進(jìn)行了研究。仿真結(jié)果顯示本文中的估計量在有限樣本下表現(xiàn)良好。隨后，本文將該方法應(yīng)用于選自滬深兩市300只高流動股票的高頻數(shù)據(jù)上，發(fā)現(xiàn)該模型有效的刻畫了中國股票市場波動的動態(tài)特征。