多元q-高斯分布统计性质探讨
首发时间:2012-11-14
摘要:本文以带随机项的微分方程组导出的一种多元函数为研究对象,验证了它是一族关于q的多尺度密度函数(多元q-高斯密度函数);然后研究了多元q-高斯密度函数的一些性质,特别是指出了在参数 β=1/2且 时它是通常我们称为的多元高斯分布密度函数,并且通过图讨论了多元q-高斯分布的一些性质,发现它较高斯分布具有尖峰厚尾性;最后推导出了多元q-高斯分布的一些数字特征(期望和协方差)。
For information in English, please click here
Discussed statistical properties of Multiple q-gaussion distribution
Abstract:this paper takes a multivariate function which derived from differential equations with random term system as the research object, verified it is gens about q multi-scale density function (multiple q - gaussian density function); And then research some properties of the multiple q - gaussian density function , especially points out that when the parameter β = 1/2 and q tend to 1 it is the multivariate gaussian distribution density function that usually we called , and discuss some properties of the multiple q - gaussian distribution through the map, find that it has peak and thick tail ; The last deduces some digital features of multiple q - gaussian distribution (expectations and covariance).
Keywords: kmultiple q - gaussian distribution Multivariate gaussian distribution Expect covariance
基金:
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
多元q-高斯分布统计性质探讨
评论
全部评论0/1000