[R-SIG-Finance] Value-at-Risk

[Adams, Zeno]

1. VaR can be estimated by several approaches. Whether the VaR model is applicable for forecasting strongly depends on the estimation method. For example, if you estimate the (univariate) VaR of a portfolio return series using GARCH, then you will have some autoregressive structure that allows you to make good 1-step-ahead forecasts.

2. You could compare the forecasted VaR with the estimated VaR at point t+1 and compute e.g. the RMSE. However, since you can only estimate the VaR but do not know the "true" VaR this may not be optimal. There is no VaR model that can forecast the first large negative returns in a series of large negative returns that you often see when plotting a return series. However, VaR models differ in how quickly they can adapt to this new situation once the first strongly negative return has been realized.

3. You might want to take a look at http://ssrn.com/abstract=1233442 where this issue is discussed in detail. In short: The traditional way of comparing models such as hit ratios or the Christoffersen (1998) test may not be optimal. If you are interested in the performance measure described in this paper I can send you the code. (At that time I programmed it in EViews. shame on me).



-----Ursprüngliche Nachricht-----
Von: r-sig-finance-bounces at stat.math.ethz.ch [mailto:r-sig-finance-bounces at stat.math.ethz.ch] Im Auftrag von Wei-han Liu
Gesendet: Dienstag, 30. Juni 2009 17:16
An: R-SIG-Finance at stat.math.ethz.ch
Betreff: [R-SIG-Finance] Value-at-Risk

Dear R-users:

Several questions please on Value-at-Risk. 

Is Value-at-Risk designed for forecasting purpose? 

I wonder if Value-at-Risk estimated by in-sample data can be used for out-of-sample forecasting? 

If in-sample Value-at-Risk is estimated by several methods, is it appropriate to do the model comparisons based on out--of-sample performance?

Wei-han Liu


      
	[[alternative HTML version deleted]]


EBS European Business School gemeinnuetzige GmbH - Sitz der Gesellschaft: Wiesbaden, Amtsgericht Wiesbaden HRB 19951 - Umsatzsteuer-ID DE 113891213 Geschaeftsfuehrer: Prof. Dr. Christopher Jahns,  Praesident; Dr. Reimar Palte,  Kanzler/CFO;  Sabine Fuchs, CMO; Aufsichtsrat: Dr. Hellmut K. Albrecht, Vorsitzender