Using Intelligent Optimization Methods to Improve the Group Method of Data Handling in Time Series Prediction

Maysam Abbod and Karishma Deshpande

School of Engineering and Design, Brunel University, West London, UK Uxbridge, UK, UB8 3PH
Maysam.Abbod@Brunel.ac.uk

Abstract. In this paper we show how the performance of the basic algorithm of the Group Method of Data Handling (GMDH) can be improved using Genetic Algorithms (GA) and Particle Swarm Optimization (PSO). The new improved GMDH is then used to predict currency exchange rates: the US Dollar to the Euros. The performance of the hybrid GMDHs are compared with that of the conventional GMDH. Two performance measures, the root mean squared error and the mean absolute percentage errors show that the hybrid GMDH algorithm gives more accurate predictions than the conventional GMDH algorithm.

Keywords: GMDH, GA, PSO, time series, prediction, finance