python 数据分析基础 day19－使用statsmodels进行逻辑回归

#使用逻辑回归预测客户流失概率

import pandas as pd
import numpy as np
import statsmodels.api as sma 

#导入数据
inputCsv='数据路径'
churn=pd.read_csv(inputCsv)


#数据预处理
#将列标题的空格替换为下划线，将引号和问号去除，标题字母变为小写
churn.columns=churn.columns.str.replace(' ','_').str.replace('\'','').str.strip('?').str.lower()
#将churn字段值中'.'删除,
churn.churn=churn.churn.str.strip('.')
#print(churn.head(5))
#新增一个字段，将churn字段转换为01编码字段
churn['churn01']=np.where(churn.churn=='True',1,0)
#对字段intl_plan及vmail_plan进行独热编码（新增虚拟变量）
intl_plan_dummy=pd.get_dummies(churn.intl_plan,prefix='intl_plan')
vmail_plan_dummy=pd.get_dummies(churn.vmail_plan,prefix='vmail_plan')
#添加常数项及生成自变量和因变量
churnInd=sma.add_constant(churn[churn.columns.difference(['intl_plan','vmail_plan','churn01','churn','state','phone','account_length','area_code'])].join(intl_plan_dummy.intl_plan_yes).join(vmail_plan_dummy.vmail_plan_yes))
churnDep=churn['churn01']


#将数据划分为训练集和测试集，训练集为第一行至倒数第10行，测试集为最后10行
churnIndTrain=churnInd.iloc[0:-10,:]
churnDepTrain=churnDep.iloc[0:-10]
churnIndTest=churnInd.tail(10)


#根据训练集训练获取模型参数
lr=sma.Logit(churnDepTrain,churnIndTrain)
result=lr.fit()
print(result.summary2())


#根据模型获取测试集结果
predictedValue=lr.fit().predict(churnIndTest)
compare=pd.DataFrame({'predictedValue':predictedValue,'actualValue':churnDep.tail(10)})
print(compare)