Stock Forecasting Model Based on Combined Fuzzy time series and genetic algorithm

Today, in the real world the forecasting of stock price time series is very important and challenging task. In recent years many fuzzy time series forecasting method have been proposed. In previous studies the fuzzy relationships were assigned with equal weights. As the fuzzy relationship were treated as equally important, therefore in forecasting each individual relationship were not reflected correctly. In proposed algorithm given in the paper we have combined fuzzy logic and genetic algorithm and uses the RMSE as fitness function to improve the forecasting accuracy. We have also recommended that different weights should be assigned to various fuzzy relationships to reflect the importance of each individual.