1

I have two numeric vectors and I would like to plot a histogram for each of them + normal curve.

First numeric vector:

dput(X)
structure(c(18.9006028896526, 15.2623176606927, 23.9827366796017, 
18.6674504871855, 33.8321828287622, 106.070218436199, 33.7827125058274, 
138.544803100033, 98.8988553851087, 84.6705010348182, 90.0070387381623, 
97.842536232733, 6.75830201534835, 24.105734944894, 18.9289005033733, 
107.837417018034, 91.2295363960887, 120.394907406909, 23.4284311509232, 
27.936658956423), .Names = c("A", "B", "C", "D", "E", "F", "G", "H", 
"I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T"))

Second numeric vector:

> dput(Y)

structure(c(4.98024718191362, 4.97784623179944, 37.54860832645, 
34.0616843614727, 146.39674720645, 17.3962674768585, 40.896642118419, 
71.7799735926384, 46.5749573881639, 39.3924128572005, 137.396714992547, 
111.856816465825, 80.8041773807388, 24.1694521970975, 15.700639434151
), .Names = c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15"))

Those vectors differ in length. I would like to apply the code similar to the one below which I used for single histogram and estimation of the normal curve:

h<-hist(X, breaks=25, col="green", xlab="Graph", 
        main="Histogram and Normal Curve", xlim = c(0, 100)) 
xfit<-seq(min(X),max(X),length=40) 
yfit<-dnorm(xfit,mean=mean(X),sd=sd(X)) 
yfit <- yfit*diff(h$mids[1:2])*length(X) 
lines(xfit, yfit, col="blue", lwd=2)

Desired output:

Something like that

Shaxi Liver
  • 1,052
  • 3
  • 25
  • 47
  • So what exactly is the question here? What is the desired output? – MrFlick Feb 29 '16 at 18:34
  • I have no idea how to put both histograms on the same graph. I mean I would like to see the bars from each vector close to each other – Shaxi Liver Feb 29 '16 at 19:37
  • What does "on the same graph" mean to you? Can you upload a sketch of what you're after? This isn't very clear. Histograms are not graphs that combine well with each other. – MrFlick Feb 29 '16 at 19:39
  • Added the desired output. Looks like a barplot... – Shaxi Liver Mar 01 '16 at 09:06

1 Answers1

2

If you want both histograms to be presented side by side, you can use the par() function:

EDIT: according to your first comment, what you actually want is a histogram with two different series... So here is the new code:

hist(X, breaks=25, col=rgb(0,1,0,alpha=.25), xlab="Graph", 
        main="Histograms and Normal Curves", xlim = c(0, 100)) 
xfit<-seq(min(X),max(X),length=40) 
yfit<-dnorm(xfit,mean=mean(X),sd=sd(X)) 
yfit <- yfit*diff(h$mids[1:2])*length(X) 
lines(xfit, yfit, col="green", lwd=2)

hist(Y, breaks=25, col=rgb(1,0,0, alpha=0.25), xlim = c(0, 100), add=TRUE) 
xfit2<-seq(min(Y),max(Y),length=40) 
yfit2<-dnorm(xfit2,mean=mean(Y),sd=sd(Y)) 
yfit2 <- yfit2*diff(h$mids[1:2])*length(Y) 
lines(xfit2, yfit2, col="red", lwd=2)

And the final plot:

Two Histograms on same graph

Key here is to make bars semi-transparent using the alpha argument to the rgb() function. You can read some other solutions on this link

Community
  • 1
  • 1
Carlos Celada
  • 131
  • 3