Computational Technology Resources - CCP

Keywords: grammatical evolution, Nikkei stock average, prediction.

Summary

Grammatical evolution (GE) is a kind of new evolutionary computation technique presented by Ryan, Collins and O'Neill [1]. It is related to the idea of genetic programming in that the objective is to find an executable program or program fragment that will achieve a good fitness value for the given objective function to be minimized.

In the GE, the translation rules from genotype (bit string) to phenotype (function or program) are defined according to the Backus-Naur form(BNF).Except for the translation rules, the algorithm is very similar to the genetic algorithm (GA). In the GA, a population of abstract representations (chromosomes or genotypes) of candidate solutions (individuals, creatures, or phenotypes) to an optimization problem evolves toward better solutions by using the genetic operators. Traditionally, solutions in the GA are represented in binary form as strings of 0s and 1s. The GE translates the strings of GA candidate solutions to the tree structure according to the BNF syntax.

In this study, three improved algorithms are presented. An original GE has two difficulties. One is related to the rule selection scheme and another is to the selection probability of candidate symbols. In order to overcome the above difficulties, improved algorithms are presented,which are named as scheme 1, 2 and 3. Numerical results show that:

the convergence properties of scheme 1 and scheme 1+3 are worse than that of the original GE, and
scheme 1+2 and scheme 1+2+3 are superior to the original GE.

The above results indicate the scheme 2 is effective for the prediction problem of the Nikkei stock average. The prediction function of the Nikkei stock average is a high-order polynomial function. Scheme 2 tends to accelerate the growth of the polynomial function. Therefore, it is concluded that scheme 2 is effective for the problem.

References

1: C. Ryan, J.J. Collins, M. O'Neill, "Grammatical evolution:Evolving programs for an arbitrary language", In "Proceedings of 1st European Workshop on Genetic Programming", Springer-Verlag, 83-95, 1998. doi:10.1007/BFb0055930

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £125 +P&P)

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Proceedings ISSN 1759-3433 CCP: 94 PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY Edited by: Paper 101 Prediction of the Nikkei Stock Average by using Advanced Grammatical Evolution T. Kuroda, H. Iwasawa and E. Kita Graduate School of Information Science, Nagoya University, Japan doi:10.4203/ccp.94.101 purchase the full-text of this paper Full Bibliographic Reference for this paper T. Kuroda, H. Iwasawa, E. Kita, "Prediction of the Nikkei Stock Average by using Advanced Grammatical Evolution", in , (Editors), "Proceedings of the Seventh International Conference on Engineering Computational Technology", Civil-Comp Press, Stirlingshire, UK, Paper 101, 2010. doi:10.4203/ccp.94.101 Keywords: grammatical evolution, Nikkei stock average, prediction. Summary Grammatical evolution (GE) is a kind of new evolutionary computation technique presented by Ryan, Collins and O'Neill [1]. It is related to the idea of genetic programming in that the objective is to find an executable program or program fragment that will achieve a good fitness value for the given objective function to be minimized. In the GE, the translation rules from genotype (bit string) to phenotype (function or program) are defined according to the Backus-Naur form(BNF).Except for the translation rules, the algorithm is very similar to the genetic algorithm (GA). In the GA, a population of abstract representations (chromosomes or genotypes) of candidate solutions (individuals, creatures, or phenotypes) to an optimization problem evolves toward better solutions by using the genetic operators. Traditionally, solutions in the GA are represented in binary form as strings of 0s and 1s. The GE translates the strings of GA candidate solutions to the tree structure according to the BNF syntax. In this study, three improved algorithms are presented. An original GE has two difficulties. One is related to the rule selection scheme and another is to the selection probability of candidate symbols. In order to overcome the above difficulties, improved algorithms are presented,which are named as scheme 1, 2 and 3. Numerical results show that: the convergence properties of scheme 1 and scheme 1+3 are worse than that of the original GE, and scheme 1+2 and scheme 1+2+3 are superior to the original GE. The above results indicate the scheme 2 is effective for the prediction problem of the Nikkei stock average. The prediction function of the Nikkei stock average is a high-order polynomial function. Scheme 2 tends to accelerate the growth of the polynomial function. Therefore, it is concluded that scheme 2 is effective for the problem. References 1 C. Ryan, J.J. Collins, M. O'Neill, "Grammatical evolution:Evolving programs for an arbitrary language", In "Proceedings of 1st European Workshop on Genetic Programming", Springer-Verlag, 83-95, 1998. doi:10.1007/BFb0055930 purchase the full-text of this paper (price £20) go to the previous paper go to the next paper return to the table of contents return to the book description purchase this book (price £125 +P&P)
Back to top	©Civil-Comp Limited 2023 - terms & conditions