金融资产收益分布的混合高斯分析

Mixture Gaussian Analysis for Chinese Securities Markets Based on Kolmogorov-Smirnov Test

本文研究了金融资产收益的混合高斯分布模型 ,给出了混合高斯分布的 Kolmogorov-Smirnov检验的方法 ,分析了金融资产收益的非高斯性及市场价格运动的有效性 .此外 ,用成分数目 K*、拟合误差 DK*n和主成分系数 p*k 描述金融资产收益的性质 。

In this paper, we present a mixture Gaussian modelling to fit finanical asset return distributions and define some measurements for indicating prices movement and then analyse the efficiency of prices movement based on Kolmogorov\|Smirnov test. The key ideals are that mixture Gaussian modelling is used to fit financial asset return distributions and assert this modelling by Kolmogorov\|Smirnov test. We have given some emperical analyses and then shown some evidences on Forex Markets and Chinese Securities Markets. It is shown that non\|Gaussality or non\|Random Walk in Chinese Securities markets is more serious than Forex Markets.