数据变换+正态检验

非正态分布-->正态：

在将非正态分布的数据转换为正态分布的数据时，首先根据偏度、峰度等情况进行判断，不同的情况可以使用对数、开根号等转换方法：

http://blog.sina.com.cn/s/blog_13ec735f50102x59u.html

 https://www.cnblogs.com/minks/p/6259883.html

http://blog.sina.com.cn/s/blog_751bd9440102vy6j.html

python-检验33品种数据是否是正态分布———————https://blog.csdn.net/qq_26948675/article/details/73648910（根据样本数选择合适的检验分布方法）；http://blog.sina.com.cn/s/blog_80828ce00101oiso.html

matlab判断是否服从正态分布：https://wenku.baidu.com/view/035d6bfadd3383c4bb4cd291.html

判断数据是否服从正态分布方法：

https://blog.csdn.net/QimaoRyan/article/details/72861387  CSDN方法

https://wenku.baidu.com/view/043521325a8102d276a22f51.html

****************https://www.cnblogs.com/webRobot/p/6760839.html************

# -*- coding: utf-8 -*-

'''

Author：Toby

QQ：231469242，all right reversed,no commercial use

 

'''

 

import scipy

from scipy.stats import f

import numpy as np

import matplotlib.pyplot as plt

import scipy.stats as stats

# additional packages

from statsmodels.stats.diagnostic import lillifors

 

group1=[2,3,7,2,6]

group2=[10,8,7,5,10]

group3=[10,13,14,13,15]

list_groups=[group1,group2,group3]

list_total=group1+group2+group3

 

 

#正态分布测试

def check_normality(testData):

    #20<样本数<50用normal test算法检验正态分布性

    if 20<len(testData) <50:

       p_value= stats.normaltest(testData)[1]

       if p_value<0.05:

           print"use normaltest"

           print "data are not normal distributed"

           return  False

       else:

           print"use normaltest"

           print "data are normal distributed"

           return True

     

    #样本数小于50用Shapiro-Wilk算法检验正态分布性

    if len(testData) <50:

       p_value= stats.shapiro(testData)[1]

       if p_value<0.05:

           print "use shapiro:"

           print "data are not normal distributed"

           return  False

       else:

           print "use shapiro:"

           print "data are normal distributed"

           return True

       

    if 300>=len(testData) >=50:

       p_value= lillifors(testData)[1]

       if p_value<0.05:

           print "use lillifors:"

           print "data are not normal distributed"

           return  False

       else:

           print "use lillifors:"

           print "data are normal distributed"

           return True

     

    if len(testData) >300: 

       p_value= stats.kstest(testData,'norm')[1]

       if p_value<0.05:

           print "use kstest:"

           print "data are not normal distributed"

           return  False

       else:

           print "use kstest:"

           print "data are normal distributed"

           return True

 

 

#对所有样本组进行正态性检验

def NormalTest(list_groups):

    for group in list_groups:

        #正态性检验

        status=check_normality(group1)

        if status==False :

            return False

             

     

#对所有样本组进行正态性检验   

NormalTest(list_groups)