A comparison of several time-series models for assessing the value at risk of shares (Q2722300)

The purpose of this paper is to assess the suitability of some standard time series models to perform a specific task in the context of recent change in banking regulations in Germany. The task is to estimate the value at risk associated with financial assets on a daily basis. The risks covered by the regulations include liquidity risks (inability to meet debts on time) and performance risks (inability to meet debts at all). Important among the latter are market risks, that is potential losses resulting from changes in market rates and prices, for example in interest rates, exchange rates, commodity prices and share prices. The investigation described in this paper is motivated by a change in the regulations regarding how the market risks are to be quantified. This paper reports on the results of an investigation of nine models that provide estimates for the first percentile of the one-step ahead forecast distribution of daily returns. To compare these models, each of them is fitted to the daily returns of 10 shares on the Frankfurt stock exchange. The data set is from the period January 3, 1974 to March 31, 1998.