Research On The Sector Trend Of The Stock Market Based On Minimum Spanning Tree

The fluctuation of our nation’s stock price in recent years shows volatility,uncertainty and randomness. Various aspects of studies of the volatility of the stockmarket and correlate structure among different stocks has been made whether from theaspect of the investor’s investment or the scholar’s academic research. However, becauseof the vagaries of the stock market, it is unreasonable for us to analysis and explain thisby the conventional method. It has an important significance to analysis the features ofthe fluctuation and the correlate structure among the different stocks from the point ofdynamic time theory.Based on the theory of minimum spanning tree, considering the structure of theminimum spanning tree, we divide nodes of the tree into three categories, includingcentral node, sub-central node and marginal node, apply this idea to the industry sector ofstock market and make an explanation of the market characteristics of these three typesof sectors. Then we generate a set of dynamic minimum spanning tree and extract threetypes of sectors to describe the total market fluctuation situation.The results show that the classification based on the structure of the minimumspanning tree has an important practical significance. The central sectors which havestable price fluctuation represent the trend of major sectors, the marginal ones which hasactive price fluctuation describe the trend of abnormal sectors. The variation of marginalsector usually plays the role of the leader, so it is speculative and profitable. In mostcases, the fluctuations structure of the central and marginal sector can make a predictionon the following trend. Three types of sectors can make a good description of the overallmarket. This description reflects the tendency and variability, eliminates a lot ofredundant information, and conduces to investment analysis.