Title :
Clustering in stock market based on fractal theory
Author :
Xiu, Zeng ; Hong, Peng ; Zhen, Zeng
Author_Institution :
Dept. of Comput. Sci. & Inf. Technol., Guangzhou Maritime Coll., Guangzhou, China
Abstract :
The K-line, which reflects the trend of stock, is fractal graphics with a stable fractal dimension; we use the stable fractal dimension as an important parameter in the research of stock cluster analysis. We have made an empirical research on the A-stock market of Shanghai. And the result shows that the same kind of stocks, which is clustered by the stock fractal theory, has an extremely similar trend. The method of stock clustering is very empirical, and the user can cluster stocks into different kinds according to the various given threshold of T. Thus, the character of the stocks can be found through analysis, and the investor will know how to operate in the stock market to avoid losing money.
Keywords :
fractals; pattern clustering; stock markets; K-line; fractal graphics; stable fractal dimension; stock fractal theory; stock market clustering; Computer graphics; Computer science; Cybernetics; Educational institutions; Euclidean distance; Fractals; Information technology; Machine learning; Relays; Stock markets; Clustering; Fractal dimension; Self-coherence function; Stock market;
Conference_Titel :
Machine Learning and Cybernetics, 2009 International Conference on
Conference_Location :
Baoding
Print_ISBN :
978-1-4244-3702-3
Electronic_ISBN :
978-1-4244-3703-0
DOI :
10.1109/ICMLC.2009.5212496
