Customized Decision Tree for Fast Multi-resolution Chart Patterns Classification

Given the advancement in algorithmic trading, the needs for real-time monitoring of patterns and execution of trades in stock exchanges become increasing important for investors. However, real-time monitoring of patterns from a vast number of markets become inefficient when tens of thousands of time series are required to be processed. In order to alleviate these problems, we propose a novel approach called Multi-resolution Chart Patterns Classification (FMCPC) based on a decision tree. In the proposed approach, a Customized Decision Tree (CDT) is built for pattern matching. CDT starts with a detailed analysis of known patterns. CDT then forms generalizations of these examples by identifying commonalities for designing decision rules. To evaluate our approach, experiments are conducted on the real datasets containing 2,527,800 data points from 19,150 stocks across top 10 Exchanges in the world. Our results reveal that FMCPC with CDT can effectively identify the chart patterns within 1 min.

[1]  Jiangling Yin,et al.  OBST-based segmentation approach to financial time series , 2013, Eng. Appl. Artif. Intell..

[2]  Tak-Chung Fu,et al.  Representing financial time series based on data point importance , 2008, Eng. Appl. Artif. Intell..

[3]  Yuqing Wan,et al.  A formal approach to chart patterns classification in financial time series , 2017, Inf. Sci..

[4]  Tak-Chung Fu,et al.  Stock time series pattern matching: Template-based vs. rule-based approaches , 2007, Eng. Appl. Artif. Intell..

[5]  Eamonn J. Keogh,et al.  Derivative Dynamic Time Warping , 2001, SDM.

[6]  Robert H. Shumway,et al.  Time series analysis and its applications : with R examples , 2017 .

[7]  Tak-chung Fu,et al.  Flexible time series pattern matching based on perceptually important points , 2001 .

[8]  Yuqing Wan,et al.  Effect of segmentation on financial time series pattern matching , 2016, Appl. Soft Comput..

[9]  Eamonn J. Keogh,et al.  A Simple Dimensionality Reduction Technique for Fast Similarity Search in Large Time Series Databases , 2000, PAKDD.

[10]  Yaojun Wang,et al.  A Case-Based Reasoning-Decision Tree Hybrid System for Stock Selection , 2016 .

[11]  Eamonn J. Keogh,et al.  Searching and Mining Trillions of Time Series Subsequences under Dynamic Time Warping , 2012, KDD.

[12]  Eamonn J. Keogh,et al.  An online algorithm for segmenting time series , 2001, Proceedings 2001 IEEE International Conference on Data Mining.

[13]  Taimur A Khan,et al.  Cointegration of International Stock Markets: An Investigation of Diversification Opportunities , 2011 .

[14]  Xueqing Li,et al.  A Novel Multi-resolution Representation for Streaming Time Series , 2017, IIKI.

[15]  C. Kirkpatrick,et al.  Technical Analysis: The Complete Resource for Financial Market Technicians , 2006 .