Matrix Exponential GARCH和近代時序模型避險績效比較與基差效果

The Performance on Hedging Models：Comparison of Matrix Exponential GARCH Model and Modern Time Series Models and Basis Effects

陳孝盈(Hsiao-Ying Chen)

指導教授 ： 李享泰

國立暨南國際大學/管理學院/財務金融學系/碩士(2008年)

摘要

本篇研究欲利用Kawakatsu （2006）所推展出多變量Matrix Exponential GARCH（ME-GARCH）來從事避險，並引用諸如BEKK-GARCH、VC-GARCH及CC-GARCH等多變量的時間序列模型從事避險績效比較，同時也加入基差與不對稱正負基差兩種效果，觀察加入兩種基差效果前後其績效改善的狀況；本文使用1998年到2007年期間中的指數期貨商品從事避險，其資料來源皆取自於DATASTREAM 資料庫；依據實證結果指出，發現在加入傳統基差效果後，大多能提升其模型之避險績效，且四個模型中以ME-GARCH模型之樣本外避險績效最佳，所以建議避險者在挑選模型以從事期貨避險時，可將ME-GARCH模型做為提升避險績效之選擇。

關鍵字

多變量GARCH ； ME-GARCH ； 動態避險 ； 基差效果

並列摘要

The purpose of this paper is to apply the Matrix Exponential multivariate Generalized Autoregressive Conditional Heteroscedasticity (ME-GARCH) proposed by Kawakatsu (2006) for futures hedging and compare the hedging effectiveness with BEKK-GARCH、VC-GARCH and CC-GARCH models under basis effects and asymmetric positive and negative spreads. This paper investigates index futures contracts collected from DATASTREAM over the period from 1998 to 2007. Based on our research, we find traditional basis effects could improve hedging effectiveness more than asymmetric positive and negative spreads. We also find ME-GARCH with basis effects outperforms BEKK-GARCH、VC-GARCH and CC-GARCH out-of-sample in terms of variance reduction. Therefore, we suggest ME-GARCH with basis effects an alternative for hedgers to improve their hedging effectiveness.

並列關鍵字

Multivariate GARCH ； Matrix Exponential GARCH ； Dynamic Hedging ； Basis effects

參考文獻

Baillie, R.T. and Myers, R.J., 1991. Bivariate GARCH Estimation of the Optimal Commodity Futures Hedge. Journal of Applied Econometrics, 6 , pp.109-124.

Bollerslev. T., 1986. Generalised Autoregressive Conditional Heteroscedasticity. Journal of Econometrics, 33, pp.307-327.

Bollerslev. T., 1990. Modeling the coherence in short-run nominal exchange rates: A multivariate generalized ARCH model. The Review of Economics and Statistics, 31, pp.498-505.

Broyden, C. G., 1970. The Convergence of a Class of Double-rank Minimization Algorithms. Journal of the Institute of Mathematics and Its Applications, 6, pp.76-90.

Dickey, D. A. and Fuller, W. A., 1981. Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica, 49, pp.1057-1072.