Evolving artificial neural networks to combine financial forecasts

Author

Harrald, Paul G. ; Kamstra, Mark

Author_Institution

Inst. of Sci. & Technol., Univ. of Manchester Inst. of Sci. & Technol., UK

Volume

1

Issue

1

fYear

1997

fDate

4/1/1997 12:00:00 AM

Firstpage

40

Lastpage

52

Abstract

We conduct evolutionary programming experiments to evolve artificial neural networks for forecast combination. Using stock price volatility forecast data we find evolved networks compare favorably with a naive average combination, a least squares method, and a kernel method on out-of-sample forecasting ability-the best evolved network showed strong superiority in statistical tests of encompassing. Further, we find that the result is not sensitive to the nature of the randomness inherent in the evolutionary optimization process

Keywords

finance; forecasting theory; genetic algorithms; neural nets; statistical analysis; encompassing; evolutionary programming; financial forecasts; forecast combination; kernel method; least squares method; naive average combination; out-of-sample forecasting ability; statistical tests; stock price volatility forecast data; Artificial neural networks; Economic forecasting; Economic indicators; Exchange rates; Genetic programming; Kernel; Least squares methods; Neural networks; Predictive models; Testing;

fLanguage

English

Journal_Title

Evolutionary Computation, IEEE Transactions on

Publisher

ieee

ISSN

1089-778X

Type

jour

DOI

10.1109/4235.585891

Filename

585891