Evaluation of Value-at-Risk
碩士 === 國立暨南國際大學 === 國際企業學系 === 91 === Value-at-Risk (VaR) models have been radically developed to measure the market risk. In this paper, we apply both hypothesis-testing and relative performance criteria to evaluate different VaR models. The results suggest that both SWARCH-L model and a...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2003
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/42661678645909130064 |
| Summary: | 碩士 === 國立暨南國際大學 === 國際企業學系 === 91 === Value-at-Risk (VaR) models have been radically developed to measure the market risk. In this paper, we apply both hypothesis-testing and relative performance criteria to evaluate different VaR models. The results suggest that both SWARCH-L model and adjusted-historical simulation model have better performance across all criteria. The strength of SWARCH approach is its efficiency to track the evolution of risk in terms of its highest correlation, only it tends to produce too few exceptions. For future researches, we suggest it may be more accurate to allow for more than two regimes or to add the GARCH term in practice.
|
|---|