Predictive Method for Exchange Rate Volatility

Abstract

The time series, used to describe the stock prices and Exchange rate, have two
characteristics: kurtosis and volatility. Actually, there are models that study kurtosis and volatility such as non lineal models: Garch models, conditional volatility and stochastic volatility models. These models are used in market risk for the forecasting of the exchange rate in the short term. The first models (Garch and conditional volatility) define volatility with its volatility from the past periods and volatility financial shocks. The second models are like Garch models but use a stochastic process knows as Wienner. This process uses simulating random walks of the exchange rate with simulating volatilities by stochastic equations. In this paper, we will analyze the performance of Garch models verses the actual models used to give forecast of the exchange rate.