# [R] Probabilities greather than 1 in HIST

David Winsemius dwinsemius at comcast.net
Thu Mar 3 17:58:28 CET 2011

```On Mar 3, 2011, at 10:29 AM, jpmaroco wrote:

> Dear David,
>
>
> I see your point. But how can I get a histogram with relative
> frequencies? If I use
>
>> plot(xhist,yhist)
>
> I get absolute frequencies in the Y axis.

I do not know of any simple setting to do what you want only within
hist.

I explained that you need to suppress the y axis ( read help(par) for
how to do that, possibly with xaxt="n"), calculate the values you
desire, then add axis labeling at the locations on the density scale
but with the probabilities you get from the calculations I illustrated.

>
> Best,
>
> Joao
>
>
>
> From: David Winsemius [via R] [mailto:ml-node+3333619-1272092771-215234 at n4.nabble.com
> ]
> Sent: quinta-feira, 3 de Março de 2011 16:01
> To: jpmaroco
> Subject: Re: Probabilities greather than 1 in HIST
>
>
>
>
> On Mar 3, 2011, at 8:03 AM, jpmaroco wrote:
>
>
>> Dear all,
>> I am a newbie in R and could not find help on this problem. I am
>> trying to
>> plot an histogram with probabilities in the y axis. This is the code
>> I am
>> using:
>>
>> #TLC uniform
>> n=30
>> mi=1; mx=6
>> nrep=1000
>> xbar=rep(0,nrep)
>> for (i in 1:nrep) {xbar[i]=mean(runif(n,min=mi,max=mx))}
>> hist(xbar,prob=TRUE,breaks="Sturges",xlim=c(1,6),main=paste("n =",n),
>>
>> The problem is that I am getting greater than 1 probabilities in the
>> Y axis?
>> Is there a way to correct this?
>
>
> Despite the argument name, which I agree suggests that probabilities
> will be plotted, what is really described in the help page is that
> densities will be plotted, and densities may be greater than 1. You
> can suppress plotting of the y-axis, calculate the probabilities for
> each of the groups returned by hist,  and then use the axes function.
>
>> xhist <- hist(xbar,breaks="Sturges",plot=FALSE)
>> yhist <- xhist\$counts/sum(xhist\$counts)
>> yhist
>  [1] 0.002 0.027 0.087 0.236 0.287 0.228 0.107 0.021 0.004 0.001
>
>
>
>
>> Joao
>>
>> --
>> View this message in context: http://r.789695.n4.nabble.com/Probabilities-greather-than-1-in-HIST-tp3333388p3333388.html
>>  <http://r.789695.n4.nabble.com/Probabilities-greather-than-1-in-HIST-tp3333388p3333388.html?by-user=t
>> >
>> Sent from the R help mailing list archive at Nabble.com.
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> and provide commented, minimal, self-contained, reproducible code.
>
>
> David Winsemius, MD
> Heritage Laboratories
> West Hartford, CT
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> and provide commented, minimal, self-contained, reproducible code.
>
>
>
>  _____
>
> discussion below:
>
> http://r.789695.n4.nabble.com/Probabilities-greather-than-1-in-HIST-tp3333388p3333619.html
>
> To unsubscribe from Probabilities greather than 1 in HIST, click
> here <http://r.789695.n4.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=3333388&code=anBtYXJvY29AZ21haWwuY29tfDMzMzMzODh8LTE1ODEwNTQwMA==
> > .
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/Probabilities-greather-than-1-in-HIST-tp3333388p3333670.html
> Sent from the R help mailing list archive at Nabble.com.
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help