Combining Hidden Markov Model and Case Based Reasoning for Time Series Forecasting

Hidden Markov Model is one of the most popular and broadly used for representation vastly structured series of data. This paper presents the application of the new approach of Hidden Markov Model and three ensemble nonlinear models to forecasting the foreign exchange rates. The proposed approach and other combination of computational intelligent techniques such as multi layer perceptron, support vector machine are compared with root mean squared error (RMSE) and Mean Absolute Error (MAE) as the performance measures. The results indicate that the new approach of Hidden Markov Model yield the best results consistently over all the currencies. and Case Based Reasoning based ensembles Based on the numerical experiments conducted, it is inferred that using the correct sophisticated ensemble methods in the computational intelligence paradigm can enhance the results obtained by the extent techniques to forecast foreign exchange rates. This suggests that the new approach of HMM is a powerful analytical instrument that is satisfactorily compared to using only the single model and other soft computing techniques for exchange rate predictions.