Fuzzy Autocorrelation Model with Fuzzy Confidence Intervals and its Evaluation
-
- Yabuuchi Yoshiyuki
- Faculty of Economics, Shimonoseki City University
-
- Kawaura Takayuki
- Department of Mathematics, Kansai Medical University
-
- Watada Junzo
- World Collaborative Innovation Center of Management Engineering
Search this article
Abstract
<p>Interval models based on fuzzy regression and fuzzy time-series can illustrate the possibilities of a system using the intervals in the model. Thus, the aim is to minimize the vagueness of the model in order to describe the possible states of the system. In the present study, we consider on an interval fuzzy time-series model based on a Box–Jenkins model, a fuzzy autocorrelation model proposed by Yabuuchi, and a fuzzy regressive model proposed by Ozawa. We examine two models by analyzing the Japanese national consumer price index and demonstrate that our approach improves the accuracy of predictions. The utility and predictive accuracy of fuzzy time-series models are validated using two concepts of fuzzy theory and statistics. Finally, we demonstrate the applicability of the fuzzy autocorrelation model with fuzzy confidence intervals.</p>
Journal
-
- Journal of Advanced Computational Intelligence and Intelligent Informatics
-
Journal of Advanced Computational Intelligence and Intelligent Informatics 20 (4), 512-520, 2016-07-20
Fuji Technology Press Ltd.
- Tweet
Details 詳細情報について
-
- CRID
- 1390282763128048512
-
- NII Article ID
- 130007673382
-
- NII Book ID
- AA12042502
-
- ISSN
- 18838014
- 13430130
-
- NDL BIB ID
- 027611869
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- NDL
- Crossref
- CiNii Articles
-
- Abstract License Flag
- Disallowed