zoukankan      html  css  js  c++  java
  • 花呗数据挖掘-破产情况分析

    我也是刚开始学习数据分析,所以直方图、柱形图、折线图这些也是挺蒙圈的,看视频学学这个花呗模型还不错

    import numpy as np
    import pandas as pd
    import matplotlib.pyplot as plt
    import matplotlib.style as psl  #绘图风格
    # % matplotlib inline  #魔法函数,在其他IDE中可能是必须的
    psl.use('seaborn-bright')
    from pylab import mpl
    mpl.rcParams['font.sans-serif'] = ['SimHei']    # 指定默认字体:解决plot不能显示中文问题
    
    #构建税率计算函数
    
    def tax(salary):
        if salary<=3500:
            tax=0
        elif salary<=(3500+1500):
            tax=(salary-3500)*0.03
        elif salary<=(3500+1500+3000):
            tax=1500*0.03+(salary-3500-1000)*0.1
        elif salary<=(3500+1000+3000+4500):
            tax=1500*0.3+3000*0.1+(salary-3500-1000-3000)*0.2
        elif salary<=(3500+1000+3000+4500+26000):
            tax=1500*0.03+3000*0.1+4500*0.2+(salary-3500-1000-3000-4500)*0.25
        elif salary<=(3500+1000+3000+4500+26000+20000):
            tax=1500*0.03+3000*0.1+4500*0.2+26000*0.25+(salary-3500-1000-3000-4500-26000)*0.3
        elif salary<=(3500+1000+3000+4500+26000+20000+25000):
            tax=1500*0.03+3000*0.1+4500*0.2+26000*0.2+20000*0.25+(salary-3500-1000-3000-4500-26000-25000)*0.35
        else:
            tax=1500*0.03+3000*0.1+4500*0.2+26000*0.2+20000*0.25+25000*0.3+(salary-3500-1000-3000-4500-26000-25000)*0.45
        return tax
    
    
    
    #构建五险一金函数
    def insurance(salary):
        if salary<21396:
            return salary*0.175
        else:
            return 3744.58
    
    #构建奖金随机函数
    def bonus(bonus_avg):
        #Series是一种类似于一维数组的对象,这里生成Series对象
        return pd.Series(np.random.normal(loc=bonus_avg,scale=200,size=120))
    #随机生成120个以平均工资为bonus_avg为正态分布的数,200为标准差,决定瘦胖,loc为均值
    # psl.use('seaborn-bright') #使用何种样式
    # print(bonus(1500))
    # plt.title('直方图')
    # plt.hist(bonus(1500),bins=30)
    #bins是一个整数,它定义了x宽度范围内的等宽面元数量,如果bin是序列,它定义了允许非均匀bin宽度的bin边缘
    # plt.show()
    
    
    #构建月净收入函数
    #净收入=月薪+奖金-五险一金-个人所得税
    def final_income(month,bonus_avg):
        df=pd.DataFrame({
            '月薪':[month for i in range(120)],
            '奖金':bonus(bonus_avg),
            '五险一金':[insurance(month) for j in range(120)],
            '个人所得税':[tax(month) for k in range(120)],
        })
        df['月净收入']=df['月薪']+df['奖金']-df['五险一金']-df['个人所得税']
        return df
    
    # result=final_income(4500,1000)
    # print(result.head())
    # result['月净收入'].iloc[:12].plot(kind='bar',figsize=(12,4))
    # plt.title('前12个月净收入')
    # plt.show()
    
    #每月支出=基本生活支出+购物支出+娱乐支出+学习支出+其他支出
    
    def expense():
        df=pd.DataFrame({
            '基本生活支出':pd.Series(np.random.randint(3000,3500,size=120)),#生成在3000-3500范围内的数值
            '购物支出':pd.Series(np.random.normal(loc=5000,scale=500,size=120)),
            '娱乐支出':pd.Series(np.random.randint(400,1200,size=120)),#生成在400-1200范围内的数值
            '学习支出':pd.Series(np.random.randint(100,500,size=120)),
            '其他支出':pd.Series(np.random.normal(loc=500,scale=40,size=120)),
        })
        df['月总支出']=df['基本生活支出']+df['购物支出']+df['娱乐支出']+df['学习支出']+df['其他支出']
        return df
    # result=expense()
    # result[['基本生活支出','购物支出','娱乐支出','学习支出','其他支出']].iloc[:12].plot.bar(figsize=(12,4),colormap='Reds_r',stacked=True)
    #柱形图长度为月总支出,因为是前几个的长度相加所得!
    # plt.show()
    
    
    #花呗还款情况分析
    #第一回合:不使用分期
    def case_a():
        income=final_income(10000,1500)['月净收入'].tolist()
        expen=expense()['月总支出'].tolist()
        saving=[0 for i in range(120)]  #月初余额
        debt=[0 for j in range(120)]  #本月需还花呗
    
        data=[]  #存储本月信息
        for i in range(120):
            money=income[i]+saving[i]-expen[i]-debt[i]   #本月剩下的钱
            if (-money)>15000:
                print('第%i个月破产了!!!'%i)
                break
            else:
                if money>=0:
                    #说明有余额,存的了钱
                    saving[i+1]=money
                    debt[i+1]=0   #负债为0
                else:
                    #说明需要用花呗借钱
                    saving[i+1]=0
                    debt[i+1]=(-money)   #需要用花呗借的钱
            data.append([income[i],expen[i],debt[i],saving[i+1],debt[i+1]])  #本月收入,支出,本月余额,本月欠款
    
        resule_a=pd.DataFrame(data,columns=['月收入','月支出','本月需还花呗','本月余额','本月欠款'])
        resule_a.index.name='月份'
        return resule_a
    
    
    #第二回合:花呗分期
    
    def case_b(n):
        income=final_income(10000,1500)['月净收入'].tolist()
        expen=expense()['月总支出'].tolist()
        saving=[0 for i in range(120)]  #月初余额
        debt=[0 for j in range(120)]  #本月需还花呗
    
        data=[]  #存储本月信息
        for i in range(120):
            money=income[i]+saving[i]-expen[i]-debt[i]   #本月剩下的钱
            if (-money)>15000:
                print('第%i个月破产了!!!'%i)
                break
            else:
                if money>=0:
                    #说明有余额,存的了钱
                    saving[i+1]=money
                    debt[i+1]=0   #负债为0
                else:
                    #说明需要用花呗借钱
                    if n==3:
                        money_pre=(abs(money)*(1+0.025))/3  #下个月要还的花呗
                        debt[i+1]=debt[i+1]+money_pre   #假设分期3个月
                        debt[i+2]=debt[i+2]+money_pre
                        debt[i+3]=debt[i+3]+money_pre
                    elif n==6:
                        money_pre = (abs(money) * (1 + 0.045)) / 6  # 下个月要还的花呗
                        debt[i + 1] = debt[i + 1] + money_pre  # 假设分期6个月
                        debt[i + 2] = debt[i + 2] + money_pre
                        debt[i + 3] = debt[i + 3] + money_pre
                        debt[i + 4] = debt[i + 4] + money_pre
                        debt[i + 5] = debt[i + 5] + money_pre
                        debt[i + 6] = debt[i + 6] + money_pre
                    elif n==9:
                        money_pre = (abs(money) * (1 + 0.065)) / 9  # 下个月要还的花呗
                        debt[i + 1] = debt[i + 1] + money_pre  # 假设分期9个月
                        debt[i + 2] = debt[i + 2] + money_pre
                        debt[i + 3] = debt[i + 3] + money_pre
                        debt[i + 4] = debt[i + 4] + money_pre
                        debt[i + 5] = debt[i + 5] + money_pre
                        debt[i + 6] = debt[i + 6] + money_pre
                        debt[i + 7] = debt[i + 7] + money_pre
                        debt[i + 8] = debt[i + 8] + money_pre
                        debt[i + 9] = debt[i + 9] + money_pre
                    else:
                        money_pre = (abs(money) * (1 + 0.088)) / 12  # 下个月要还的花呗
                        debt[i + 1] = debt[i + 1] + money_pre  # 假设分期12个月
                        debt[i + 2] = debt[i + 2] + money_pre
                        debt[i + 3] = debt[i + 3] + money_pre
                        debt[i + 4] = debt[i + 4] + money_pre
                        debt[i + 5] = debt[i + 5] + money_pre
                        debt[i + 6] = debt[i + 6] + money_pre
                        debt[i + 7] = debt[i + 7] + money_pre
                        debt[i + 8] = debt[i + 8] + money_pre
                        debt[i + 9] = debt[i + 9] + money_pre
                        debt[i + 10] = debt[i + 10] + money_pre
                        debt[i + 11] = debt[i + 11] + money_pre
                        debt[i + 12] = debt[i + 12] + money_pre
                    saving[i + 1] = 0
            data.append([income[i],expen[i],debt[i],saving[i+1],debt[i+1]])  #本月收入,支出,本月余额,本月欠款
    
        resule_a=pd.DataFrame(data,columns=['月收入','月支出','本月需还花呗','本月余额','本月欠款'])
        resule_a.index.name='月份'
        return resule_a
    
    #一万次模拟
    def similar():
        month=[]
        for p in range(10000):
            month.append(case_b(12).index.max()+1)  #添加最大索引再加一,也就是破产月份
    
        month=pd.DataFrame(month,columns=['月份'])
        month.plot.hist(figsize=(12,4))
        plt.show()
    
    #使用花呗情况比较
    r1=case_a()['本月欠款']
    r2=case_b(3)['本月欠款']
    r3=case_b(6)['本月欠款']
    r4=case_b(9)['本月欠款']
    r5=case_b(12)['本月欠款']
    result_b=pd.DataFrame({'不分期':r1,'分期3月':r2,'分期6月':r3,'分期9月':r4,'分期12月':r5},
                          columns=['不分期','分期3月','分期6月','分期9月','分期12月'])
    #生成折线图
    result_b.plot.line(alpha=0.8,style='--',colormap='Accent',figsize=(12,4),use_index=True,legend=True)
    plt.title('不同情况下的破产情况')
    plt.show()
    

      

  • 相关阅读:
    How to using X++ code achieve copying records
    How to using x++ code achieve Lookup files list with entire folder
    Using X++ code create and Read XML file.
    Using X++ Code Create master table form
    When you are finished renamed the Item number Jobs
    Visual Studio.NET 简介
    Visual C++中的异常处理浅析
    最常见的20种VC++编译错误信息
    开发WDM型的USB设备驱动程序
    C++中内存管理
  • 原文地址:https://www.cnblogs.com/fodalaoyao/p/10692985.html
Copyright © 2011-2022 走看看