カイゾウジ， タイセイ   KAIZOJI, Taisei   海蔵寺　大成    所属   国際基督教大学教養学部 アーツ・サイエンス学科    職種   教授
言語種別	英語
発行・発表の年月	2018/07
形態種別	研究論文
査読	査読あり
標題	Computational intelligence methods for data mining of causality extent in time series
執筆形態	共著
掲載誌名	International Journal of Computational Science and Engineering
掲載区分	国外
出版社・発行元	INDERSCIENCE
巻・号・頁	16(4),pp.411-418
著者・共著者	L. Pichl and T. Kaizoji
概要	We adopt the support vector machine (SVM) and artificial neural network (ANN) for causality rate extraction. The dataset records all details of the futures contracts on the commodity of gasoline traded in Japan. By sampling the tick data at 1 min, 5 min, 10 min, 30 min, 1 hour and 1-day scales, we derive time series of varying causal degrees. Trend predictions are computed by using the SVM binary classifier trained on 66.6% of the data using a five-step-back moving window which samples the log-returns as the predictor data. From the testing data, we extract varying rates of causality degree, starting from the borderline of 50% up to the order of 60% in rare cases. The trend prediction analysis is complemented by the ANN method with four hidden layers. Overall, the market of the gasoline futures in Japan is found to be rather close to the efficient market hypothesis in comparison with other commodities markets.
researchmap用URL	https://www.inderscience.com/info/inarticle.php?artid=93782