Applications of the Moving Average of nth -Order Difference Algorithm for Time Series Prediction

Authors:
Yang Lan;Daniel Neagu
Affiliations:
Department of Computing, University of Bradford, Bradford, BD7 1DP, UK;Department of Computing, University of Bradford, Bradford, BD7 1DP, UK
Venue:
ADMA '07 Proceedings of the 3rd international conference on Advanced Data Mining and Applications
Year:
2007

Citing 4
Cited 0

Matrix computations (3rd ed.)

Matrix computations (3rd ed.)
Noisy Time Series Prediction using Recurrent Neural Networks and Grammatical Inference

Machine Learning
Time series forecasting: Obtaining long term trends with self-organizing maps

Pattern Recognition Letters - Special issue: Artificial neural networks in pattern recognition
Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Currently, as a typical problem in data mining, Times Series Analysis and Prediction are facing continuously more applications on a wide variety of domains. Huge data collections are generated or updated from science, military, financial and environmental applications. Prediction of the future trends based on previous and existing values is of a high importance and various machine learning algorithms have been proposed. In this paper we discuss results of a new approach based on the moving average of the nth-order difference of limited range margin series terms. Based on our original approach, a new algorithm has been developed: performances on measurement records of sunspots for more than 200 years are reported and discussed. Finally, Artificial Neural Networks (ANN) are added for improving the precision of prediction by addressing the error of prediction in the initial approach.