Research on seruential data pattern discovery method for categorization of stock price fluctuation patterns and estimation of fluctuation factors

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K00161
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Multimedia database
• Research Institution: Tokyo Polytechnic University
• Principal Investigator: Udagawa Yoshihisa 東京工芸大学, 工学部, 教授 (50582574)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: テクニカル分析 / 株価予測 / ローソク足チャート / チャートパターン / 数理モデル

Outline of Final Research Achievements

In finance, technical analysis is a technique that aims to forecast the future direction of stock prices through the study of past market data primarily on price and volume. Candlestick charting represented by "Sakata five methods" has been originally developed in Japan and then become popular around the world. While candlestick charting is widely used to predict short-term price movements, there is still no consistent conclusion among academic researches whether this approach leads to profits. In this research, stock price predictions based on candlestick chart patterns are examined using stock price indicators of Japan and the US stock markets for about 10 years.
As a result, observing the patterns that have been proposed in "Sakata five methods" with taking into account the stock price fluctuations up to the next day, it was confirmed that the proposed stock price predictions are expected to produce a profit.

Free Research Field

データマイニング

Academic Significance and Societal Importance of the Research Achievements

貯蓄から投資にシフトすることで日本経済の活性化が試みられているが、2016年度の日本銀行の調査によれば、家計に占める投資比率はむしろ減少傾向にある。精度の高い株価変動予測法を開発し、投資リスクを低減させることが、投資比率を高めるための鍵になっている。
本研究で開発した株価変動の予測方法は、ビッグデータ分析を先導するものと評価され、国内外の学術会議や雑誌でも論文として採択された実績がある。社会的には、投資リスクを軽減することを通して投資への関心を高める礎を提供した意義がある。