python相关系数显著性检验矩阵_python散点图及相关系数矩阵计算和相关性验证

python散点图及相关系数矩阵计算

# -*- coding:utf-8 -*-

# __author__ = "LQ"

import matplotlib.pyplot as plt

import pandas as pd

import scipy.stats as stats

#自定义方法调用，上一篇有该方法

from DataAnalysis.TeachingModel.dbc import dbcConnect

def main(sql):

df = dbcConnect.selectDf(sql)

print(df)

# pearson相关系数矩阵

corr = df.corr()

print(corr)

#计算某一个元素相关

print(df.corr()[u'avg_getscore'])

# 输出结果第一个值为pearsonr相关系数，

# 第二个为p-value，所以这里Guba列和Value值是显著相关的

pearsonr = stats.pearsonr(df['avg_total_afterrequire_num'], df['avg_getscore'])

print(pearsonr)

# 散点图矩阵

pd.plotting.scatter_matrix(df, figsize=(8, 8),

c='k',

marker='+',

diagonal='hist',

alpha=0.8,

range_padding=0.1)

plt.show()

if __name__ == '__main__':

sql = '''SELECT

avg_total_afterrequire_num,

avg_total_beforerequire_num,

avg_total_homework_num,

avg_actual_homework_num,

avg_total_exam_num,

avg_actual_exam_num,

avg_evaluate_num,

avg_question_answer_num,

avg_note_num,

avg_error_num,

avg_login_log_num,

avg_question_times,

avg_getscore

FROM

t_study_behavior_class

WHERE

avg_getscore > 0

AND avg_getscore < 100'''

main(sql)

自定义方法dbcConnect

在下面博客中

/qq_30868737/article/details/103995174