Forecasting Value-at-Risk Using Conditional Volatility Models: Evidence from Tehran Stock Exchange

Abstract

In this paper, we investigate the performance of parametric ARCH
class models to forecast out-of-sample VaR for two portfolios of
Tehran Stock Exchange (TSE) companies (Market portfolio and a
portfolio of 50 liquid companies), using a number of distributional
assumptions and sample sizes at low and high confidence levels. We
find, first, that leptokurtic distributions are able to produce better oneday-
ahead and 10-day-ahead VaR forecasts; second, the choice of
sample size is important for the accuracy of the forecasts.

Keywords