t检验的代码

#coding=utf-8
"""
@author=wanggang
data=2020.10.30
"""
from scipy.stats import ttest_1samp
import numpy as np
from scipy.stats import ttest_ind, norm, f
import numpy as np
def read():
   sample=[]
"""
def ftest(s1, s2):
   '''F检验样本总体方差是否相等'''
   print("Null Hypothesis:var(s1)=var(s2)，α=0.05")
   F = np.var(s1) / np.var(s2)
   v1 = len(s1) - 1
   v2 = len(s2) - 1
   p_val = 1 - 2 * abs(0.5 - f.cdf(F, v1, v2))
   print(p_val)
   if p_val < 0.05:
      print("Reject the Null Hypothesis.")
      equal_var = False
   else:
      print("Accept the Null Hypothesis.")
      equal_var = True
   return equal_var


def ttest_ind_fun(s1, s2):
   '''t检验独立样本所代表的两个总体均值是否存在差异'''
   equal_var = ftest(s1, s2)
   print("Null Hypothesis:mean(s1)=mean(s2)，α=0.05")
   ttest, pval = ttest_ind(s1, s2, equal_var=equal_var)
   if pval < 0.05:
      print("Reject the Null Hypothesis.")
   else:
      print("Accept the Null Hypothesis.")
   return pval

"""
#np.random.seed(42)
#s1 = norm.rvs(loc=1, scale=1.0, size=20)
#s2 = norm.rvs(loc=1.5, scale=0.5, size=20)
#s3 = norm.rvs(loc=1.5, scale=0.5, size=25)

#ttest_ind_fun(s1, s2)
#ttest_ind_fun(s2, s3)
print("原假设：假设非专业均值和总体的相等")
print("备择假设:假设非专业的均值不等于总体的均值")
print("Null Hypothesis:μ=μ0=30，α=0.05")
ages = [25,36,15,40,28,31,32,30,29,28,27,33,35]
t = (np.mean(ages)-30)/(np.std(ages,ddof=1)/np.sqrt(len(ages)))#这里的30 为总体的平均值 ddof=1 是计算的标准差。np.sqrt(len(ages))这个是

ttest,pval = ttest_1samp(ages,30)#计算p 值
#print(t,ttest)
if pval < 0.05:
   print("拒绝原假设，备择假设成立。存在显著性差异")
else:
   print("接收原假设，不存在显著性差异")