[2P4-J-2] Machine learning: industries and finance

[2P4-J-2-04] A study of hybrid prediction model considering risk reduction for economic indicators forecasting

〇Kodai Ogihara¹, Shoichi Urano¹ (1. Meiji University)

Keywords:time series data, multiple regression model, artificial neural network

In this paper, we have proposed the hybrid prediction model considering risk reduction for economic indicators forecasting. The proposed hybrid prediction model consists of multiple regression model and artificial neural network. We apply the proposed hybrid prediction model to Nikkei 225 futures price. Based on the simulation results, we report the result that it is possible to avoid risk by suppressing the maximum prediction error rate and standard deviation while keeping prediction accuracy in the proposed hybrid prediction model with Nikkei 225 futures price.