An Adaptive Radial Basis Function Neural Network Glowworm Swarm Optimization for Time Series Forecasting

Abstract

The past three decades have witnessed active research using a class of Artificial Neural Network models (ANN), the Radial Basis Function Neural Networks, to forecast time series. Many techniques for forecasting time series using Radial Basis Function Neural Networks (RBFNN) have been proposed and developed in literature. The major challenges in RBFNN lie in the optimization of its full parameters: the number and location of cluster centres as well as the output weights. To address these challenges, this study adapted the Clustering Analysis based on Glowworm Swarm Optimization (CGSO) algorithm to obtain a modified Clustering Analysis based on Glowworm Swarm Optimization (CGSOm) algorithm for solving the clustering problem. Adaptation was achieved by incorporating a mechanism that determines the sensor range of the CGSO efficiently and automatically, and modifying the glowworm initialization method. For the weight optimization, the Bioluminescence Swarm Optimization algorithm (BSO) was adopted, making it the first time it will be applied in training the weights of the RBFNN. Other training algorithms tested include Conjugate Gradient Descent (CGD), Gradient Descent (GD) and Particle Swarm Optimization algorithm (PSO). Stock price and currency exchange rate data were used to train the combinations of models developed. The results obtained from the training showed that the CGSOm-CGD RBFNN gave best forecasting accuracy by yielding lowest error values; followed by the CGSOm-BSO RBFNN that gave relatively similar error values. Hence, two new training methodologies for time series forecasting resulted from this study; they are the CGSOm-BSO RBFNN and the CGSOm-CGD RBFNN.