python实现均值类指标差异分析-Mann-Whitney U-test(完整代码)

1

# 导入需要的库

2

from warnings import simplefilter

3

simplefilter(action='ignore', category=FutureWarning)

4

import pandas as pd

5

import numpy as np

6

import scipy.stats as stats

7

8

# =============读取数据===========

9

def Read_data(file):

10

dt = pd.read_csv(file)

11

dt.columns = ['age', 'sex', 'chest_pain_type', 'resting_blood_pressure', 'cholesterol', 'fasting_blood_sugar','rest_ecg', 'max_heart_rate_achieved', 'exercise_induced_angina', 'st_depression', 'st_slope','num_major_vessels', 'thalassemia', 'target']

12

data = dt

13

return data

14

15

# ===========数据清洗==============

16

def data_clean(data):

17

# 重复值处理

18

print('存在' if any(data.duplicated()) else '不存在', '重复观测值')

19

data.drop_duplicates()

20

# 缺失值处理

21

print('不存在' if any(data.isnull()) else '存在', '缺失值')

22

data.dropna() # 直接删除记录

23

data.fillna(method='ffill') # 前向填充

24

data.fillna(method='bfill') # 后向填充

25

data.fillna(value=2) # 值填充

26

data.fillna(value={'resting_blood_pressure': data['resting_blood_pressure'].mean()}) # 统计值填充

27

# 异常值处理

28

data1 = data['resting_blood_pressure']

29

# 标准差监测

30

xmean = data1.mean()

31

xstd = data1.std()

32

print('存在' if any(data1 > xmean + 2 * xstd) else '不存在', '上限异常值')

33

print('存在' if any(data1 < xmean - 2 * xstd) else '不存在', '下限异常值')

34

# 箱线图监测

35

q1 = data1.quantile(0.25)

36

q3 = data1.quantile(0.75)

37

up = q3 + 1.5 * (q3 - q1)

38

dw = q1 - 1.5 * (q3 - q1)

39

print('存在' if any(data1 > up) else '不存在', '上限异常值')

40

print('存在' if any(data1 < dw) else '不存在', '下限异常值')

41

print(data)

42

return data

43

44

#=============mannwhitneyu检验======================

45

def mannwhitneyu_age(data):

46

data1=data[['age','target']]

47

print(data1['age'].values)

48

print(np.mean(data1['age'].values), np.std(data1['age'].values))

49

data_age_Y=data1[(data1['target'] ==1)]

50

print(len(data_age_Y))

51

print(data_age_Y['age'].values)

52

print(np.mean(data_age_Y['age'].values), np.std(data_age_Y['age'].values))

53

data_age_N=data1[(data1['target'] ==0)]

54

print(len(data_age_N))

55

print(data_age_N['age'].values)

56

print(np.mean(data_age_N['age'].values), np.std(data_age_N['age'].values))

57

MWU=stats.mannwhitneyu(data_age_Y['age'].values, data_age_N['age'].values, alternative='two-sided')

58

print(MWU)

59

60

#============主函数==============

61

if __name__=="__main__":

62

data1=Read_data("F:\数据杂坛\\0504\heartdisease\Heart-Disease-Data-Set-main\\UCI Heart Disease Dataset.csv")

63

data1=data_clean(data1)

64

mannwhitneyu_age(data1)