文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 经管营销 > 金融/投资 > GARCH—GED Model Based—AGA for Financial Risk Measure
侵权投诉

GARCH—GED Model Based—AGA for Financial Risk Measure

GARCH—GED Model Based—AGA for Financial Risk Measure

【Abstract】GARCH model under the assumption of generalized error distribution (GED) is used to obtain the daily-volatility of log-return. The parameters are estimated by adaptive genetic algorithm. The tests of VaR and CVaR show that GARCH-GED model is valid for financial risk measure.

【Key words】GARCH; Generalized error distribution; Adaptive genetic algorithm; VaR; CVaR.

1 GARCH-GED Model and AGA

GARCH model has been widely used for financial risk measure. A standard GARCH (1,1) model is as follows [1]:

2 Data and Results

The return is the close price of Shindex from 1997-12-17 to 2010-12-17, is left-skewed distribution (-0.0956) and higher kurtosis (6.9853). Q-Q plot shows that the normality hypothesis is refused. The convergence of AGA is in Figure 2.

Figure 1 Q-Q plot

Figure 2 The results of AGA

The estimation for GARCH-GED are as follows:ω=4.58e-6,α=0.096,β=0.891>0.8,α+β=0.987<1,

v=1.232<2. These indicate GARCH is stationary and the Shindex is fatted-tail and higher kurtosis.

In order to clearly observe the fitting degree of different distributions, we plot the probability density curves and sample empirical density curve in Figure 1. Figure 2 shows us the degree of left tail fitting for shindex.

Figure 3 Several distributions for shindex

Figure 4 Left tail fitting for shindex

It can be seen that the curves of GED and student’s-t are almost in coincidence with the kernel density estimation curve, but the fitting effect of normal distribution is poor.

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email