Hybrid approach for time series forecasting based on ANFIS and Fuzzy Cognitive Maps

In this paper we consider a hybrid approach to forecasting time series using neuro-fuzzy prediction models and Fuzzy Cognitive Maps. Main idea of proposed approach is hybridization two different ways for time series forecasting. We can make quantitative and qualitative forecast. In addition, we describes the different approaches to learning and optimization of the network, such as the methods of particle swarm, evolutionary methods, as well as variants of the hybridization of these methods. Also, in a comparison of the results of the prediction for example, one of the indicators of the State Program of Development of Science and Technology, is forecast schedule.