Estimating VaR based on EVT used with GARCH and Wavelet analysis

Bibliographic Information

Other Title
  • ウェーブレット解析とGARCHを用いたEVTによるVaR推定

Search this article

Description

Estimating VaR (Value at Risk) that is used with conventional EVT (Extreme Value Theory), historical data that exceeds the sample standard deviation at the observation period is considered to be the tail distribution. After the tail distribution is estimated, VaR is Estimated. For conventional EVT model, the beginning point of the tail distribution, in a word, the threshold is not so referred. However, it is one of important problem in EVT to estimate the threshold correctly. In this paper, the volatility is estimated by wavelet analysis applying to the profit margin time series, and using GARCH for the time series after wavelet analysis. The estimated volatility is assumed to be a threshold and VaR is estimated with EVT. VaR used with proposing EVT model is smaller than VaR used with conventional EVT, and showed the result of passing statistical binomial test so far.

Journal

  • IPSJ SIG technical reports

    IPSJ SIG technical reports 2007 (128), 159-162, 2007-12-20

    Information Processing Society of Japan (IPSJ)

Details 詳細情報について

  • CRID
    1572261552198905216
  • NII Article ID
    110006594842
  • NII Book ID
    AA12055912
  • ISSN
    09196072
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top