A Bias-Corrected Rate-Optimal Estimator of the Integrated Covariance of Security Returns with Serially Dependent Noise
2010/02/02
研討會日期 : 2010-02-02
時間 : 14:00
主講人 : Mr. Shinsuke Ikeda
地點 : B110
演講者簡介 : Mr. Shinsuke Ikeda為Ph.D. in Economics，Boston University of (May 2010 expected)。其主要研究領域為Econometrics、Finance。現正申請本所研究職務中。
演講摘要 : A bias-corrected non-parametric estimator of the integrated covariance matrix of security returns is proposed. It is constructed using a linear combination of two realized kernels. In addition to its simplicity, my estimator has desirable statistical properties including consistency, asymptotic normality and the best parametric rate of convergence in the presence of serially dependent microstructure noise. A transformation using the Cholesky decomposition guarantees that the estimator is positive semi-definite in finite samples without altering its asymptotic properties. This transformation allows me to show that non-linear functions of the estimated matrix, such as the hedge ratio, enjoy the same asymptotic properties. Simulations show that my estimator has much smaller bias than currently available estimators with only a mild increase in variance, so that the overall mean squared error is also smaller. In an application to intra-daily data of S&P 500 spot and futures prices, my method delivers estimates of hedge ratios with an 80% reduction in bias assessed in relation to the daily benchmark compared with the realized kernel estimator.
