Log in

+34 977 759833sigef@urv.cat

Artificial intelligent based time series forecasting of stock prices using digital filters

A. Sfetsos. Imperial College

C. Siriopoulos. University of Macedonia


The aim of the paper is the analysis of the sequential characteristics of the Athens Stock Exchange general index (ASE) using the time series metho-dology based on artificial intelligent techniques. The applied models include the Feed Forward Neural Network trained with the efficient Levenberg - Marquardt optimization algorithm, the Adaptive Neuro-Fuzzy Inference Sys-tem as well as traditional linear regression and ARIMA models for comparison. All these approaches are initially used for the short-term fore-casting of the series, providing an insight into the forecasting capabilities of each model. The analysis of the spectral characteristics of the series indicated the presence of strong persis-tence or alternatively that the models do not differ significantly from a random walk. This observation was also cemen-ted by the forecasting results of the developed models. The proposed approach is based on the application of low-pass digital filters on the series and the employment of the formerly mentioned models for the prediction of the created series. The filtered series contains a lower amount of noise and can be viewed as an alternative trend indication of the original series.

Keywords: stock prices, forecasting, neural networks, ANFIS, filters.

You must be a member to download the full article.

Log in or Sign up

Cron Job Starts