Non-Linear Markov Modelling Using Canonical Variate Analysis: Forecasting Exchange Rate Volatility

We report on a novel forecasting method based on nonlinear Markov modelling and canonical variate analysis, and investigate the use of a prediction algorithm to forecast conditional volatility. In particular, we assess the dynamic behaviour of the model by forecasting exchange rate volatility. It is found that the nonlinear Markov model can forecast exchange rate volatility significantly better than the GARCH(1,1) model due to its flexibility in accommodating nonlinear dynamic patterns in volatility, which are not captured by the linear GARCH(1,1) model.