|
Author(s): |
Jiping Yang, School of Economics and Management, Beihang University, Beijing, China, 100191
Chunhui Zhang, School of Economics and Management, Beihang University, Beijing, China, 100191
Lu Yuan, School of Economics and Management, Beihang University, Beijing, China, 100191 |
|
Abstract: |
In the paper, we perform empirical research on VaR using two kinds of methods in the semiparametric models and non-parametric model: estimating function method and historical simulation approach on the daily return data of the Shanghai Composite Index from January 5, 2009 to September 30, 2011 for two confidence levels. The results show that compared with the historical simulation method, the VaR obtained by the estimating function method can not only pass through the test with high accuracy, but also have broader applicability. After that, we improve the original historical simulation method and find that compared to the original historical simulation method, the improved one can get more accurate VaR.
|