China stock market correlation mining algorithm based on FP-Tree

Xu Tiansheng, Qin Aiming and S

Abstract

The research is based on FP-Tree algorithm for mining frequent phase sets that use hash tables and conditional probability formula to get stock correlation rules between the ups and downs. Ten different China A stocks of a quarter in 2013 are used for testing the FP tree method developed in this paper. The results show that there are correlations in different categories in China A stocks using FP-Tree algorithm. The correlations found in this paper can be used for the investors to make a decision. The algorithm proposed can be applied to stock arbitrage, long-term buying and selling contracts, and composing investment portfolios.