A Range-in-DCC Model for Correlation Coefficient Estimation
2004/11/30
研討會日期 : 2004-11-30
時間 : 15:00
主講人 : Prof. Chou, Ray Yeutien
地點 : B棟110室
演講者簡介 : Prof. Chou, Ray Yeutien為加州大學聖地牙哥分校經濟學博士(1988)。
現為本所副研究員。
其主要領域為計量經濟、財務金融、時間數列分析及財務計量分析。
演講摘要 : This paper proposes a Range-in-DCC model, which is an extension of Engle’s (2002) Dynamic Conditional Correlation (DCC) model. The efficiency of the range data in volatility estimation is documented in Parkinson (1980), Alizadeh, Brandt and Diebold (2001), Brandt and Jones (2002) and Chou (2004), among others. It is hence natural to consider the implication of this result in the estimation of multivariate GARCH models. In the DCC model, the conditional correlation coefficients are estimated by a dynamic model for the product of the pair-wise return series each normalized by their conditional standard deviations. The conditional standard deviation is calculated by using a univariate GARCH for the return series.
In this paper, we use the Conditional Autoregressive Range (CARR) model of Chou (2004), as an alternative to the univariate GARCH in the DCC first step estimation procedure. The substantial gain in efficiency in the volatility estimation can induce efficiency gain in the estimation of the time-varying correlation coefficient. For comparison we estimate the DCC model as a benchmark to gain insights into the difference of these methods. We use two data sets for our empirical analyses: the daily stock indices of S&P500 and Nasdaq, the DJ-30 and Nasdaq. Both in-sample and out-of-sample results indicate strong support for our model in terms of the precision in estimating the correlations and covariances.
