Robust prediction of stock indices using PSO based adaptive linear combiner

The present paper employs a particle swarm optimization (PSO) based adaptive linear combiner for efficient prediction of various stock indices in presence of strong outliers in the training data. The connecting weights of the model are updated by minimizing the Wilcoxon norm of the error vector by PSO. The short and long term prediction performance of the new model is evaluated with test data and the results obtained are compared with those obtained from the conventional PSO based model. It is in general observed that the proposed model is computationally more efficient, prediction wise more accurate and more robust against outliers in training set compared to those obtained by standard PSO based model.