Risk modeling using the Extreme Value Theory. Application of Ecompassing Method and Combined Forecasts to VaR quantification.

Ewa Ratuszny 

Warsaw School of Economics, Al. Niepodległości 162, Warsaw 02-554, Poland

Abstract

In the research we compare VaR methods/models for FX instruments. We consider Extreme Value Theory, implied quantile model and ecompassing method to achieve more accurate forecasts of risk measure using simultaneously information from historical time series and information regarding market’s expectation of future risk. For risk measurement we examine also four methods of combining forecasts: simple average combining, unrestricted linear combination, a weighted averaged combining and weighted averaged combining using exponential weighting. 

 

Related papers
  1. Influence of robust estimation on volatility forecast

Presentation: Poster at 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych", by Ewa Ratuszny
See On-line Journal of 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych"

Submitted: 2015-09-21 19:55
Revised:   2015-09-21 19:55