A Hybrid Forecast of Exchange Rate based on ARFIMA,Discrete Grey-Markov, and Fractal Kalman Model

Abstract: We propose a hybrid forecast based on extended discrete grey Markov and variable dimension Kalman model and show that our hybrid model can improve much more the performance of forecast than traditional grey Markov and Kalman models. Our simulation results are given to demonstrate that our hybrid forecast method combined with degree of grey incidence are better than grey Markov and ARFIMA model or Kalman methods.