[R] Multiple Histograms in R

[prateek pande]

Hi,

I have a data as mentioned below(at the bottom)

Now out of that data i have to create multiple histograms in a single view
in R. On that histogram i need on x -axis binned data with Breaks 10 and
on y axis event rate . Here churn is dependent variable.


*for example, for mou_mean , on x -axis on histogram i need Bins(mou_mean)
and on y - axis in need Churn%age. *
*Bins(mou_mean)*

*Churn %age*
23-43 0.23%
33-53 0.5%
43-63 0.3%
53-73 0.4%
63-83 0.7%
83-103 0.8%

Please help


*mou_mean*

*totalmrc_mean*

*rev_range*

*mou_range*

*Churn*

23

24

25

27

1

45

46

47

49

1

43

44

45

47

1

45

46

47

49

0

56

57

58

60

0

67

68

69

71

1

67

68

69

71

0

44

45

46

48

1

33

34

35

37

0

90

91

92

94

1

87

88

89

91

1

76

77

78

80

1

33

34

35

37

1

44

45

46

48

1

[[alternative HTML version deleted]]

______________________________________________
R-h...@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[Rui Barradas]

Hello,

Maybe package psych, function multi.hist is what you want.

https://cran.r-project.org/web/packages/psych/index.html

And don't post in HTML, your data is unreadable.

Hope this helps,

Rui Barradas

[Bert Gunter]

Also, *if* this is homework, don't post at all, as this list is not
for doing homework (though, for better or worse, sometimes such help
is provided).


-- Bert


Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )

[Jim Lemon]

Hi Prateek,
There is some difficulty with including the empty categories in the
factors generated. I couldn't get these even with drop=FALSE, so I had
to go through the "xtab" function. You can do it with the "table"
function in the base package, but it is a little more trouble. See if
this is what you want>

ppdat<-read.table(text="mou_mean,totalmrc_mean,rev_range,mou_range,Churn
23,24,25,27,1
45,46,47,49,1
43,44,45,47,1
45,46,47,49,0
56,57,58,60,0
67,68,69,71,1
67,68,69,71,0
44,45,46,48,1
33,34,35,37,0
90,91,92,94,1
87,88,89,91,1
76,77,78,80,1
33,34,35,37,1
44,45,46,48,1",
sep=",",header=TRUE)
ppdat$mou_mean_cut<-cut(ppdat$mou_mean,breaks=seq(23,103,10),include.lowest=TRUE)
ppdat$totalmrc_mean_cut<-cut(ppdat$totalmrc_mean,breaks=seq(23,103,10))
ppdat$rev_range_cut<-cut(ppdat$rev_range,breaks=seq(23,103,10))
ppdat$mou_range_cut<-cut(ppdat$mou_range,breaks=seq(23,103,10))
library(prettyR)
ppx<-xtab(Churn~mou_mean_cut,ppdat)
mou_mean_agg<-100*ppx$counts[2,]/colSums(ppx$counts)
mou_mean_agg[is.nan(mou_mean_agg)]<-0
ppx<-xtab(Churn~totalmrc_mean_cut,ppdat)
totalmrc_mean_agg<-100*ppx$counts[2,]/colSums(ppx$counts)
totalmrc_mean_agg[is.nan(totalmrc_mean_agg)]<-0
ppx<-xtab(Churn~rev_range_cut,ppdat)
rev_range_agg<-100*ppx$counts[2,]/colSums(ppx$counts)
rev_range_agg[is.nan(rev_range_agg)]<-0
ppx<-xtab(Churn~mou_range_cut,ppdat)
mou_range_agg<-100*ppx$counts[2,]/colSums(ppx$counts)
mou_range_agg[is.nan(mou_range_agg)]<-0
ppmat<-matrix(c(mou_mean_agg,totalmrc_mean_agg,rev_range_agg,
mou_range_agg),nrow=4,byrow=TRUE)
library(plotrix)
barp(ppmat,col=rainbow(4),main="Multiple histogram",ylim=c(0,105),
names.arg=levels(ppdat$mou_mean_cut),ylab="Percent churn")
legend(2.5,107,c("mou_mean","totalmrc_mean","rev_range","mou_range"),
fill=rainbow(4))

Jim

[Hasan Diwan]

Prateek,
I'm shocked this isn't answered previously, but you can try the par command
(mfrow and mfcol parameters) and par(mfrow=n, mfcol=m) will generate n
plots per row and m rows per column. For subsequent questions, please do a
search through the archives before asking. -- H

--
OpenPGP:
https://sks-keyservers.net/pks/lookup?op=get&search=0xFEBAD7FFD041BBA1
If you wish to request my time, please do so using
http://bit.ly/hd1ScheduleRequest.
Si vous voudrais faire connnaisance, allez a
http://bit.ly/hd1ScheduleRequest.

<https://sks-keyservers.net/pks/lookup?op=get&search=0xFEBAD7FFD041BBA1>Sent
from my mobile device
Envoye de mon portable

[prateek pande]

HI Hasan,

Thanks for sharing the solution. Really appreciate it.

But i was reading somewhere that we cannot use par with ggplot 2 . we can
only use grid extra to have multiple plots in a single view.

Is it right?

Regards
Prateek

[Ulrik Stervbo]

Hi Prateek,

maybe facet_* with ggplot is what you are looking for

HTH
Ulrik