python筆記：6.3.1.2獨立樣本比例之差假設檢驗（例6-9雜誌讀者差異性）

 

# -*- coding: utf-8 -*-
"""
Created on Fri Jun 21 11:08:34 2019

@author: User
"""

# 《Python數據分析基礎》中國統計出版社

import numpy as np
from scipy import stats
import pandas as pd
import statsmodels.api as sm

magzine = pd.read_csv(u'data\\ch6\\magzine.csv',encoding = "gbk")
print(magzine.head())

'''
掛上標籤便於閱讀
'''
magzine['name']=magzine['name'].astype('category')
magzine['name'].cat.categories=['Fashion','Cosmetic']
magzine['name'].cat.set_categories=['Fashion','Cosmetic']

magzine['gender']=magzine['gender'].astype('category')
magzine['gender'].cat.categories=['Male','Female']
magzine['gender'].cat.set_categories=['Male','Female']

print("\n 先統計一下兩種雜誌的女性讀者人數")
female=magzine[magzine['gender']=='Female']['name'].value_counts()
print(female)

print("\n 兩種雜誌的總人數")
magzines=magzine['name'].value_counts()
print(magzines)

'''
參數1：成功的數目
參數2：樣本量
'''
print(sm.stats.proportions_ztest(np.array(female),
                                 np.array(magzines),
                                 value=0.3,
                                 alternative='smaller',
                                 prop_var=False))