用卡方统计量的Q-Q图检验法探讨三元数据的正态性检验问题我要分享

Using Q-Q graph test method of chi square statistics to discuss the normality test of three metadata

关注次数: 448

下载次数: 0

文件大小: 1.25 kB

代码分类: 其他

开发平台: matlab

下载需要积分: 2积分

版权声明:如果侵犯了您的权益请与我们联系,我们将在24小时内删除。

代码描述

中文说明: 用卡方统计量的Q-Q图检验法探讨三元数据的正态性检验问题。 评价正态性的方法,如果:1、大约有一半的马氏距离d小于或等于q(0.5),即这些马氏距离中一半或一半以上落在轮廓线之内; 2、画出马氏距离和与之对应的卡方百分位数的点(sort_d,qRec)的图形近似是一条通过原点且斜率为1的直线,那么表明是三元正态的。


English Description:

&Using Q-Q graph test of chi square statistics, this paper discusses the normality test of three metadata. &If: 1. About half of the Mahalanobis distance D is less than or equal to Q (0.5), that is, half or more of these Mahalanobis distances fall within the contour line; 2. Draw the points of Mahalanobis distance and the corresponding chi square percentile (sort)_ d. If the graph of qrec is approximately a straight line passing through the origin with a slope of 1, it means that it is Trivariate normal.


代码预览

相关推荐