chaotic time series prediction using the kohonen algorithm.docVIP

下载本文档

4
0
约1.52万字
约 7页
2016-03-05 发布于广东
举报
版权申诉

chaotic time series prediction using the kohonen algorithm.doc

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

chaotic time series prediction using the kohonen algorithm

Chaotic Time Series Prediction using the Kohonen Algorithm Luis Monzón Benítez1, Humberto Brito Santana1, Daniel Travieso Pedroso1, Eloy Ortiz Hernández1, Marcelo Chacón Reyes1 and Diana Pedreira Iparraguirre2. University of Camaguey. Department of Physics. Circunvalación Norte Km 5.5 CP 74650 Camaguey, Cuba. Telf: (53-322)61019 Fax: (53-322)61126 e-mail: nolineal@reduc.cmw.edu.cu Medical University “Carlos J. Finaly”. Camaguey, Cuba. Contact Author: Luis Monzón Benítez. e-mail: lmonzon@reduc.cmw.edu.cu Topic Areas: Artificial neural networks. Adaptive and Intelligent Control. 1. Introduction. In the last decades many different techniques have been developed for the analysis and prediction of time series, ranging from the well-known Box and Jenkins models[1] to the latest neural based ones[2,3,4]. The most used approach consists in considering the time series as a realization of a stochastic process and to apply the different methods devised in the framework of Statistics to obtain a model for the process[1,3,5,6]. Today, with the latest results concerning deterministic chaos, even if the data look random apparently it is important to consider the possibility that the time series was generated by a low order nonlinear deterministic dynamical system[7,8,9]. One of the techniques developed within the Chaos Theory formalism for the analysis of time series is deterministic nonlinear prediction[9]. In this paper the use of a Kohonen network as a component of one deterministic nonlinear prediction algorithm is suggested. The algorithm was originally proposed by Lorenz[10] and later modified by Ikeguchi and Aihara[9] and is known as the Modified Method of Analogues (MMA). The paper is organized as follows: In section 2 the MMA is explained and in the next section the proposed prediction algorithm is introduced. In section 4 some experiments are conducted to evaluate the performance of the algorithm and the results are discussed. Finally, the main conclusions are highligh