上市公司股利分配倾向研究:基于数据挖掘方法的分析

股利分配是上市公司的一项十分重要的财务决策。而影响股利分配的因素有许多，国内现有文献一般是采用多元统计回归建模方法对其进行实证分析。本文应用数据挖掘中的决策树方法寻找股利分配的原因，这是文章的创新点。从决策树来看，现金股利分配不仅受股本结构、公司规模、当年可分配利润、成长机会及经营风险等客观财务因素的影响，而且与控股股东性质存在显著的影响，且国家控股股东分配倾向低于民营控股股东。

Empirical Analysis of Dividend Distribution Tendency of Listed Companies:Based on the Methods of Data Mining

Dividend distribution of listed companies is a very important financial decision - making. Many domestic literature generally use multivariate statistical regression model to analyze the factors which affect the dividend distribution.In this paper,the decision tree in data mining is used to find the reasons for dividend distribution, which is the innovation of the article. From the tree view,the distribution of cash dividends is not only affected by the capital structure, company size, profits allocated forthe year,growth opportunities and business risk, but also the nature of the controlling shareholder. The results show that the tendency of the national controlling shareholder is lower than the private.