A Survey of Forecasting Preprocessing Techniques using RNs

In this paper we make a survey of various preprocessing techniques including the statistical method for volatile time series forecasting using Regularization Networks (RNs). These methods improve the performance of Regularization Networks i.e. using Independent Component Analysis (ICA) algorithms and filtering as preprocessing tools. The preprocessed data is introduced into a Regularized Artificial Neural Network (ANN) based on radial basis functions (RBFs) and the prediction results are compared with the ones we get without these preprocessing tools, with the high computational effort method based on multidimensional regularization networks (MRN) and with the Principal Component Analysis (PCA) technique.