The Markov switching multi-fractal model of asset returns : estimation and forecasting of dynamic volatility with multinomial specificaitons

The Makov-Switching Multifractal Model(MSM) is a new model for the time series of retuns in finance. It can generate various degree of long range dependence in different powers of returns. We extend this model with trinomial - and dronomial specification. Generalized method of moment(GMM) and maximum likelihood(ML) are used to estimate the parameters of the model. We use the estimates in forecasting financial volatility. Comparing the MSM forecasts with those derived from GARCH and FIGARCH models yields results in favor of the new model