pingouin是基于Pandas和numpy开发的Python3统计包。主要统计功能有

• 方差分析
• 多元线性回归
• 中介效应分析
• 卡方检验
• Q-Q图
• 贝叶斯因子
• 信效度检验
• 等等

安装

pip3 install pingouin

快速上手

构造实验数据x，y

import numpy as np

#控制代码每次随机状态保持一致
np.random.seed(666)

n=30
mean= [4,5]
cov = [(1, 0.6), (0.6, 1)]

x, y = np.random.multivariate_normal(mean, cov, n).T

x

array([3.04817645, 2.54387965, 4.56033188, 4.40504338, 3.77876203,
       3.87177128, 3.4546112 , 4.47317551, 5.23133856, 5.40273745,
       5.19344217, 3.37061786, 3.23980982, 2.85574177, 4.67728276,
       4.31935242, 4.39440207, 3.87458876, 4.91426293, 3.13673286,
       3.73459839, 4.18708647, 5.48558345, 3.7066784 , 3.73400287,
       3.49664637, 3.95954844, 2.61545452, 5.11352964, 5.62666503])

y

array([4.47747109, 4.35695696, 5.46239455, 4.56091782, 4.07534588,
       4.03904897, 3.79549165, 5.06121364, 5.71635355, 6.60772697,
       6.94890455, 5.13347618, 5.41207983, 3.38254684, 5.49705058,
       5.93394729, 4.65224366, 4.59491971, 5.17926604, 4.25844527,
       5.72809738, 5.14997732, 5.27606588, 4.94570454, 6.02889647,
       5.85451666, 4.90231286, 4.69242625, 4.69367432, 6.71644528])

import matplotlib.pyplot as plt

plt.hist(x, bins=10)

plt.hist(y, bins=10)

1. T检验

import pingouin as pg

pg.ttest(x, y)

2. 皮尔森相关

pg.corr(x, y)

3.鲁棒检验

#添加一个异常值
x[5] = 18
#使用Shepherd's pi correlation
pg.corr(x, y, method="shepherd")

4. 数据正态性检验

pg.normality(x)

pg.normality(y)

pg.multivariate_normality(np.column_stack((x, y)))

(True, 0.6257634649268228)

5. Q-Q plot

import numpy as np
import pingouin as pg

np.random.seed(666)

x = np.random.normal(size=50)
ax = pg.qqplot(x, dist='norm')

6. 单因素方差分析

# 读取数据
df = pg.read_dataset('mixed_anova')
df.sample(10)

# Run the ANOVA
aov = pg.anova(data=df, 
               dv='Scores',  #因变量
               between='Group', 
               detailed=True)
aov

7. 重复测量方差分析

pg.rm_anova(data=df, 
            dv='Scores', 
            within='Time', 
            subject='Subject', 
            detailed=True)

8. 有交互作用的双因素方差分析

# Compute the two-way mixed ANOVA and export to a .csv file
aov = pg.mixed_anova(data=df, 
                     dv='Scores', 
                     between='Group', 
                     within='Time',
                     subject='Subject', 
                     correction=False, 
                     effsize="np2")
pg.print_table(aov)

=============
ANOVA SUMMARY
=============

Source          SS    DF1    DF2     MS      F    p-unc    np2      eps
-----------  -----  -----  -----  -----  -----  -------  -----  -------
Group        5.460      1     58  5.460  5.052    0.028  0.080  nan
Time         7.628      2    116  3.814  4.027    0.020  0.065    0.999
Interaction  5.167      2    116  2.584  2.728    0.070  0.045  nan

9. 多元线性回归

pg.linear_regression(data[['X', 'Z']], data['Y'])

10. 中介效应分析

pg.mediation_analysis(data=data, x='X', m='Z', y='Y', seed=42, n_boot=1000)

Pingouin与Pandas

pandas.DataFrame可直接使用Pingouin的很多统计方法，例如

import pingouin as pg

# Example 1 | ANOVA
df = pg.read_dataset('mixed_anova')
df.anova(dv='Scores', between='Group', detailed=True)

# Example 2 | Pairwise correlations
data = pg.read_dataset('mediation')
data.pairwise_corr(columns=['X', 'M', 'Y'], covar=['Mbin'])

# Example 3 | Partial correlation matrix
data.pcorr()

pandas.DataFrame支持的pingouin统计方法有:

• pingouin.anova()
• pingouin.ancova()
• pingouin.rm_anova()
• pingouin.mixed_anova()
• pingouin.welch_anova()
• pingouin.pairwise_ttests()
• pingouin.pairwise_tukey()
• pingouin.pairwise_corr()
• pingouin.partial_corr()
• pingouin.pcorr()
• pingouin.rcorr()
• pingouin.mediation_analysis()