Combining Clonal Selection Algorithm and Gene Expression Programming for Time Series Prediction

Author

Litvinenko, V.I. ; Bidyuk, P.I. ; Bardachov, J.N. ; Sherstjuk, V.G. ; Fefelov, A.A.

Author_Institution

Kherson Nat. Tech. Univ., Kyiv

fYear

2005

fDate

5-7 Sept. 2005

Firstpage

133

Lastpage

138

Abstract

Dynamic system identification algorithm is developed using the basic mechanisms of clonal selection and idea of a new evolutionary computing paradigm - gene expression programming. On the basis of the algorithm developed a computer based system is proposed for making decisions relevant to forecasting of single variable and multivariate time series. The results of computing experiments achieved with the system developed show high quality of short and medium period forecasts.

Keywords

forecasting theory; genetic algorithms; mathematics computing; time series; clonal selection algorithm; dynamic system identification algorithm; evolutionary computing; gene expression programming; multivariate time series forecasting; single variable time series forecasting; time series prediction; Algorithm design and analysis; Diversity reception; Dynamic programming; Gene expression; Genetic expression; Genetic programming; Immune system; Nonlinear dynamical systems; Software algorithms; Time series analysis; artificial immune systems; clonal selection algorithm; gene expression programming; time series;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 2005. IDAACS 2005. IEEE

Conference_Location

Sofia

Print_ISBN

0-7803-9445-3

Electronic_ISBN

0-7803-9446-1

Type

conf

DOI

10.1109/IDAACS.2005.282956

Filename

4062107