generate data

# -*- coding: utf-8 -*-
"""
Created on Mon Sep 25 22:47:47 2017

@author: RUI54
"""

import random
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime

#generate user, user number from 1 to 10000
region = []
month = []
offer = []
acc_num = []
revenue = []
discount = []
unknown = []
population = {'Longgang':184*0.6,'Yantian':21*0.3,'Baoan':330*0.6,'Longhua':140*0.4,'Futian':116*0.65,'Luohu':86*0.35,'Nanshan':90*0.7}
loc_proporation = [0.1,0.08,0.2,0.08,0.07,0.18,0.15,0.06,0.03,0.05]   #各套餐在每个区域的占比
monthly_change1 = [0,0,-0.025,0,0,0.02,0.025,0,0,-0.01]   #第一年各套餐每个月变化趋势
monthly_change2 = [0,0,-0.02,0,0,0.01,0.03,0,0,-0.01]   #第二年各套餐每个月变化趋势
monthly_change3 = [0,0,-0.01,0,0.03,0.02,0.025,0,-0.02,-0.05]   #第三年各套餐每个月变化趋势
discount_region = {'Longgang':0.7,'Yantian':0.5,'Baoan':0.8,'Longhua':0.5,'Futian':0.6,'Luohu':0.65,'Nanshan':0.8}#the discount influence in different regions
offer_fee = [68,78,68,38,98,88,98,78,38,158]
for i in range(1,37):
    for ii in range(1,11):
        for iii in ['Longgang','Yantian','Baoan','Longhua','Futian','Luohu','Nanshan']:
            if i<10:
                month.append('2015/0'+str(i)+'/01')
            elif i>=10 and i<13:
                month.append('2015/'+str(i)+'/01')
            elif i>=13 and i<22:
                month.append('2016/0'+str(i-12)+'/01')  #month
            elif i>=22 and i<25:
                month.append('2016/'+str(i-12)+'/01')
            elif i>=25 and i<34:
                month.append('2017/0'+str(i-24)+'/01')
            else:
                month.append('2017/'+str(i-24)+'/01')
            
            offer.append('offer'+str(ii))         #offer
            region.append(iii)                    #region
            
            Discount = random.randint(1,10)
            if Discount<=5:
                dis = 0
                discount.append('null')
            elif Discount>5 and Discount<9:
                dis = 0.05
                discount.append('5%')
            else:
                dis = 0.10
                discount.append('10%')
            
            Unknown = random.uniform(-0.03,0.03)
            unknown.append(Unknown)
                            
            #各区域各套餐各月份使用人数
            if i<13:
                account_number = 10000*(population[iii]*(loc_proporation[ii-1]+random.uniform(0,0.03)))*(1+(i-1)*(monthly_change1[ii-1]+random.uniform(0,dis)*discount_region[iii]+random.uniform(-0.01,0.01)))*(1+Unknown)
            elif i>=13 and i<25:
                account_number = 10000*(population[iii]*(loc_proporation[ii-1]+random.uniform(0,0.03)))*(1+(i-1)*(monthly_change2[ii-1]+random.uniform(0,dis)*discount_region[iii]+random.uniform(-0.01,0.01)))*(1+Unknown)
            else:
                account_number = 10000*(population[iii]*(loc_proporation[ii-1]+random.uniform(0,0.03)))*(1+(i-1)*(monthly_change3[ii-1]+random.uniform(0,dis)*discount_region[iii]+random.uniform(-0.01,0.01)))*(1+Unknown)
            
            if account_number>0:
                account_number = account_number
            else:
                account_number = random.uniform(500,2000)    
            acc_num.append(account_number)
            Revenue = account_number*(offer_fee[ii-1]*(1-dis*discount_region[iii])+random.randint(-10,10))
            revenue.append(Revenue)
            
                      
dic_data = {'Month':month, 'Product':offer,'Region':region,
            'Account number':acc_num, 'Revenue':revenue, 'Discount':discount, 'Unknown factor':unknown}

df = pd.DataFrame(dic_data)
df['Account number'] = df['Account number'].astype('int')
df['Revenue'] = df['Revenue'].astype('int')


df.to_csv('generate_new_data.csv')