HW5

# coding: utf-8

'''

作业要求：

1、不同类型卡的持卡人的性别对比

2、不同类型卡的持卡人在办卡时的平均年龄对比

3、不同类型卡的持卡人在办卡前一年内的平均帐户余额对比

4、不同类型卡的持卡人在办卡前一年内的平均收入对比

'''

#########################################################

import os

import pandas as pd

os.chdir(r'D:\HW5')

loanfile = os.listdir()

createVar = locals()

for i in loanfile:

    if i.endswith("csv"):        

        createVar[i.split('.')[0]] = pd.read_csv(i, encoding = 'gbk')

        print(i.split('.')[0])

#########################################################

# In[]

#1、不同类型卡的持卡人的性别对比

data1=pd.merge(card,disp,on='disp_id')

data_con1=pd.merge(data1,clients,on='client_id')

data_con1=data_con1.rename(columns={'type_x':'type'})

gender_cross_raw=pd.crosstab(data_con1['type'],data_con1['sex'])

gender_cross=gender_cross_raw.div(gender_cross_raw.sum(1),axis=0)

from pylab import mpl

mpl.rcParams['font.sans-serif'] = ['SimHei']  # 指定默认字体

mpl.rcParams['axes.unicode_minus'] = False  # 解决保存图像是负号'-'显示为方块的问题

gender_cross.plot(kind='bar',stacked=True)

#########################################################

# In[]

#2、不同类型卡的持卡人在办卡时的平均年龄对比

from datetime import datetime

data_con1['birth_date']=pd.to_datetime(data_con1['birth_date'])

data_con1['age']=datetime.today().year-data_con1['birth_date'].dt.year

import seaborn as sns

from pylab import mpl

mpl.rcParams['font.sans-serif']=['SimHei']

mpl.rcParams['axes.unicode_minus']=False

sns.boxplot('type','age',data=data_con1)

#########################################################

# In[]

#3、不同类型卡的持卡人在办卡前一年内的平均帐户余额对比

data_con2=trans[['account_id','date','amount','balance']].merge(disp[['disp_id','account_id']],on='account_id')\

.merge(card[['disp_id','issued','type']],on='disp_id')

data_con2['balance']=data_con2['balance'].str.strip('$').str.replace(',','').astype('int')

data_con2['issued']=pd.to_datetime(data_con2['issued']).dt.date

data_con2['date']=pd.to_datetime(data_con2['date']).dt.date

data_con3=data_con2[(data_con2['issued']-data_con2['date']).dt.days<365]

#print(data_con3.head())

import seaborn as sns

from pylab import mpl

mpl.rcParams['font.sans-serif']=['SimHei']

mpl.rcParams['axes.unicode_minus']=False

sns.boxplot('type','balance',data=data_con3)

#########################################################

# In[]

#4、不同类型卡的持卡人在办卡前一年内的平均收入对比

data_con3['amount']=data_con3['amount'].str.strip('$').str.replace(',','').astype('int')

#print(data_con3.sample())

import seaborn as sns

from pylab import mpl

mpl.rcParams['font.sans-serif']=['SimHei']

mpl.rcParams['axes.unicode_minus']=False

sns.boxplot('type','amount',data=data_con3)