test

lend_club 全球最大的P2P平台。 
此文章基于R语言做简单分析。

1. rm(list=ls()) #清除变量
2. gc() #释放内存

• step1 
  考虑到后续分析 
  将数据导入sqlserver,用到SSIS 
  如图

 
 **此处有坑

• step2 
  连接sqlserver,并将数据读入R。

1. plot
   

   lend_club4<- lend_club3%>%
     group_by(m,y)%>%
     summarise(total_m=sum(sumamount))
   
   lend_club4
   head(lend_club4)
   Source: local data frame [6 x 3]
   Groups: m [2]
   
         m     y   total_m
     (chr) (chr)     (dbl)
   1    01  2008  32256329
   2    01  2009  28523635
   3    01  2010  63082946
   4    01  2011 171186425
   5    01  2012 297667575
   6    02  2008  20596688
   

   　　

   　　

• step3

1. library(ggplot2)
3. qplot(date_1,sumamount,data=lend_club1,geom="line") # 每天贷款金额的时序图

1. p<-qplot(month_day,sumamount,data=lend_club1)
2. p+facet_wrap(~year) #2007-2012 期间每日的贷款金额

1. library(tidyr)
2. library(dplyr)
3. lend_club2<-separate(lend_club1,date_1,c("y","m","d"),sep="-")
4. head(lend_club2)
5. sumamount y m d year month_day
6. 1 2000 2007 05 26 2007 05/26
7. 2 47400 2007 05 27 2007 05/27
8. 3 23900 2007 05 28 2007 05/28
9. 4 121050 2007 05 29 2007 05/29
10. 5 87500 2007 05 30 2007 05/30
11. 6 46500 2007 05 31 2007 05/31

1. lend_club3<-unite(lend_club2,"y_m",y,m,sep="-",remove = F)
2. head(lend_club3)
3. sumamount y_m y m d year month_day
4. 1 2000 2007-05 2007 05 26 2007 05/26
5. 2 47400 2007-05 2007 05 27 2007 05/27
6. 3 23900 2007-05 2007 05 28 2007 05/28
7. 4 121050 2007-05 2007 05 29 2007 05/29
8. 5 87500 2007-05 2007 05 30 2007 05/30
9. 6 46500 2007-05 2007 05 31 2007 05/31
11. qplot(m,sumamount,data=lend_club3,geom=c("boxplot")+facet_wrap(~year) #2007~2012年每月贷款金额的箱线图

1. lend_club4<- lend_club3%>%
2. group_by(m,y)%>%
3. summarise(total_m=sum(sumamount))
5. lend_club4
6. head(lend_club4)
7. Source: local data frame [6 x 3]
8. Groups: m [2]
10. m y total_m
11. (chr) (chr) (dbl)
12. 1 01 2008 32256329
13. 2 01 2009 28523635
14. 3 01 2010 63082946
15. 4 01 2011 171186425
16. 5 01 2012 297667575
17. 6 02 2008 20596688

1. 折线图 分面
2. p<-qplot(m,total_m,data=lend_club4)+geom_smooth(aes(group=y,colour=y),method = "lm")

折线图 分面

1. p<-qplot(m,total_m,data=lend_club4)+geom_smooth(aes(group=y,colour=y))

1. p+facet_wrap(~y)

1. lend<-read.csv("C:\\Users\\liyi\\Desktop\\lend_club.csv")
2. lend1<-read.csv("C:\\Users\\liyi\\Desktop\\lend_club.csv",header = F)
3. lend1<-lend1[-1,]
4. head(lend1)
5. lend1<-lend1[,c(1,3,9)]
6. myvar<-c("amount","year","employment")
7. names(lend1)<-myvar
8. head(lend1)
9. str(lend1)
10. lend1$amountnew<-as.numeric(as.character(lend1$amount))
12. library(sqldf)
14. lend2<-sqldf('select sum(V1),V3,V9
15. from lend1
16. group by V3,V9')
17. q<-qplot(employment,amountnew,data = lend1,geom=c("boxplot"),colour=lend1$employment)+facet_wrap(~year)
18. q<- q+theme(axis.text.x=element_text(angle=90,hjust=1,colour="black"),legend.position='none')
19. q<- q+scale_y_continuous(limits = c(0, 100000))
20. q