Python大数据：信用卡逾期分析

# -*- coding:utf-8 -*-
# 数据集成

import csv
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt  

#客户信息
basicInfo = pd.DataFrame.from_csv('datas/basicInfo_train.csv', header=0, sep=',', index_col=0, parse_dates=True, encoding=None, tupleize_cols=False, infer_datetime_format=False)
#历史还款记录
historyInfo = pd.DataFrame.from_csv('datas/history_train.csv', header=0, sep=',', index_col=0, parse_dates=True, encoding=None, tupleize_cols=False, infer_datetime_format=False)
#历史逾期情况
defaultInfo = pd.DataFrame.from_csv('datas/default_train.csv', header=0, sep=',', index_col=0, parse_dates=True, encoding=None, tupleize_cols=False, infer_datetime_format=False)
combineInfo = pd.concat([basicInfo,historyInfo,defaultInfo],axis=1)

#查看前10条数据
combineInfo[:10]

#性别分析
gender = combineInfo.groupby('SEX')['Default'].mean().reset_index()
plt.xticks((0,1),(u"Male",u"Female"))
plt.xlabel(u"Gender")
plt.ylabel(u"Counts")
plt.bar(gender.SEX,gender.Default,0.1,color='green')
plt.show()

#教育程度与default值的相关性分析
edu = combineInfo.groupby('EDUCATION')['Default'].mean()
plt.plot(edu)
plt.show()

#婚姻状况分析
marriage = combineInfo.groupby('MARRIAGE')['Default'].mean().reset_index()
plt.bar(marriage.MARRIAGE,marriage.Default,0.5,color='green')
plt.show()