javiferfer

import numpy as np
import seaborn as sns 
import pandas as pd
from pandas.plotting import scatter_matrix
import matplotlib.pyplot as plt 

from scipy import stats
from scipy.stats import norm

from sklearn.datasets import fetch_california_housing

javiferfer

from scipy import stats

X1 = stats.norm.rvs(loc=10, scale=2, size=30, random_state=0)
X2 = stats.norm.rvs(loc=10, scale=2, size=30, random_state=1)
X3 = stats.norm.rvs(loc=14, scale=2, size=30, random_state=2)

print(stats.ttest_ind(X1, X2))
print(stats.ttest_ind(X1, X3))

javiferfer

t_values_array = []

for i in range(100000):
    X1 = stats.norm.rvs(loc=10, scale=2, size=30, random_state=(i+1))
    X2 = stats.norm.rvs(loc=10, scale=2, size=30, random_state=2*(i+1))
        
    mean_X1 = np.mean(X1)
    mean_X2 = np.mean(X2)
    std_X1 = np.std(X1)
    std_X2 = np.std(X2)

javiferfer

t_test_function(rvs1, alpha=0.05, no_test=100)
t_test_function(rvs2, alpha=0.05, no_test=100)

bonferroni_correction_function(rvs1, alpha=0.05, no_test=100)
bonferroni_correction_function(rvs2, alpha=0.05, no_test=100)

bonferroni_holm_correction_function(rvs1, alpha=0.05, no_test=100)
bonferroni_holm_correction_function(rvs2, alpha=0.05, no_test=100)

sidak_correction_function(rvs1, alpha=0.05, no_test=100)

javiferfer

def sidak_correction_function(rvs, alpha, no_test):
    FWER = 1-(1-alpha)**(1/no_test)
    alpha_sidak = 1-(1-FWER)**(1/no_test)
    
    counter = 0
    for i in range(no_test):
        rvs_random = stats.norm.rvs(loc=5, scale=10, size=1000, random_state=i+1)

        statistic, pvalue = stats.ttest_ind(rvs, rvs_random, equal_var=False)

javiferfer

def bonferroni_holm_correction_function(rvs, alpha, no_test):
    pvalue_test = []
    for i in range(no_test):
        rvs_random = stats.norm.rvs(loc=5, scale=10, size=1000, random_state=i+1)

        statistic, pvalue = stats.ttest_ind(rvs, rvs_random, equal_var=False)
        pvalue_test.append(pvalue)
        
    pvalue_test_sorted = sorted(pvalue_test, key=float)
    

javiferfer

def bonferroni_correction_function(rvs, alpha, no_test):
    alpha_bonferroni = alpha/no_test
    
    counter = 0
    for i in range(no_test):
        rvs_random = stats.norm.rvs(loc=5, scale=10, size=1000, random_state=i+1)

        statistic, pvalue = stats.ttest_ind(rvs, rvs_random, equal_var=False)

        if pvalue <= alpha_bonferroni:

javiferfer

def t_test_function(rvs, alpha, no_test):
    counter = 0
    for i in range(no_test):
        rvs_random = stats.norm.rvs(loc=5, scale=10, size=1000, random_state=i+1)

        statistic, pvalue = stats.ttest_ind(rvs, rvs_random, equal_var=False)

        if pvalue <= alpha:
            counter = counter + 1

javiferfer

rvs1 = stats.norm.rvs(loc=5, scale=10, size=1000, random_state=0)
rvs2 = stats.norm.rvs(loc=6.5, scale=8, size=1000, random_state=0)

javiferfer

import numpy as np
import matplotlib.pyplot as plt
from scipy import stats