Research Of Stock Markets Index Sery Based On Complex Network And Data Mining

Chinese stock market has become an important part of the securities industry andfinancial industry and many scholars have proposed a variety of methods to study stockmarket laws. This paper analyzes the Shanghai Composite Index, using complex networkand data mining technology. Different from the previous methods, the article analyzes theShanghai Composite Index from the two dimensions of time and space.The article converts close price of the Shanghai Composite Index from October six2006to December eleven2012, getting the time series and the stock yield series. At thesame time, the concept of adjustment factor has been introduced, we dynamicallyanalysis the features of the time sequence and the yield sequence. Conduct the weighteddirected graph using the time series. Using the k-nuclear analysis for this network, wegain the relationships of the stock market with the nodes. For the yield sequence, weconsider the yield distribution and classify different yield distributions usingmultidimensional scale and minimum spanning tree methods.The results show that the method of k-nuclear plays an important role in the wholenetwork. And the k-nuclear has divided the nodes into three categories of normal points,transition points and abnormal points. The three kinds of nodes are closely related withthe trend of stock market. Rise and fall yield sequence distribution consistent withconventional yields distribution, has the characteristics of peak and thick tail and obeysthe asymmetric Laplace distribution. But the yield sequences gradually deviate from theasymmetric Laplace distribution under the different adjusting factors. Classifying theyield sequences with the methods of multidimensional scale and minimum spanning tree,we get six types of yield distributions.