python统计分析-双样本T检验

#!/usr/bin/env python
# -*- coding:utf-8 -*-

# <editable>

def execute():
    # <editable>
    '''
    载入模块
    '''
    from scipy.stats import ttest_ind, norm
    import pandas as pd
    from sqlalchemy import create_engine
    '''
    连接数据库
    '''
    engine = create_engine('mysql+pymysql://root:123123qwe@127.0.0.1:3306/analysis')
    '''
    选择目标数据
    '''
    # 生成数据

    # params = {
    #     "col1": "",
    #     "col2": "",
    # }
    # inputs = {"table": '纯随机性检验'}
    # data_sql = 'select ' + params['col1'] + ',' + params['col2'] + ' from ' + inputs['table']
    # data_in = pd.read_sql_query(data_sql, engine)
    # print(data_in)


    col1 = norm.rvs(loc=5, scale=10, size=500)
    col2 = norm.rvs(loc=5, scale=10, size=500)
    '''
    双样本t检验
    '''
    # col1 = data_in[params['col1']]
    # col2 = data_in[params['col2']]
    # p = ttest_ind(col1, col2)[1]
    p = ttest_ind(col1, col2)[1]
    '''
    ttest_ind(equal_var=False)
    
    equal_var : bool, optional
        If True (default), perform a standard independent 2 sample test that assumes equal population variances [R263]. 
        If False, perform Welch’s t-test, which does not assume equal population variance [R264].
    '''
    data_out = ''
    if (p < 0.05):
        data_out += '双样本t检验结果'
        data_out += '检验结果'
        data_out += "p值为：" + str(p) + ",认为两者总体均值不同"
    else:
        data_out += '双样本t检验结果'
        data_out += '检验结果'
        data_out += "p值为：" + str(p) + ",无充分证据证明两者总体均值不同"

    '''
    生成报告
    '''
    print(data_out)
# </editable>

if __name__ == '__main__':
    execute()