[R] Help needed: Fraction for Histogram > 1 ???

[Thomas Lumley]

On Mon, 22 Jun 2009, charles78 wrote:

>
> It is not the case as you described.  In any case, the total area should be 1
> and labeled fraction on y axis should be far less than 1, since I have more
> than 10000 data points.  I also test differerent bin size by change the
> break.

Try reading Greg's response again.  Then calculate the area of one of the bars 
in your histogram, remembering that the area of a rectangle is not the same as 
its height, but is the width times height.  If the area of the bar is greater than 
1, then you can report a problem.

       -thomas


> I draw the graph using only 1 group, the same result was obtained.
>
> Any othe suggestion?
>
> Charles.
>
> Greg Snow-2 wrote:
>>
>> When freq=FALSE then the y axis is not the proportion in each group (what
>> I am assuming you mean by fraction), but rather is scaled so that the
>> total area of the histogram is 1 (making comparing to theoretical
>> densities easier).  If all the data values are between 0 and 1, then the
>> height of at least one bar needs to be >= 1 for the total area to equal 1.
>>
>> If you want the y-axis to show relative frequency (proportion, fraction,
>> etc.), then you either need to plot the y-axis yourself or use a different
>> function than 'hist'.
>>
>> --
>> Gregory (Greg) L. Snow Ph.D.
>> Statistical Data Center
>> Intermountain Healthcare
>> greg.snow at imail.org
>> 801.408.8111
>>
>>
>>> -----Original Message-----
>>> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
>>> project.org] On Behalf Of charles78
>>> Sent: Monday, June 22, 2009 10:08 AM
>>> To: r-help at r-project.org
>>> Subject: [R] Help needed: Fraction for Histogram > 1 ???
>>>
>>>
>>> I have been trying to draw histogram for my manscript and found some
>>> strange
>>> things that I could not figure out why.
>>>
>>> Using the same code listed below I have successfully draw histograms
>>> for a
>>> few figures with fraction labeled on Y axis less than 1 (acturally
>>> between 0
>>> to 0.1).  But one dataset gives the Y axis label 0 to 5 as fraction.
>>> This
>>> is not true, as fraction are less than 1, although the value
>>> distribution on
>>> the figure seems to me is right.
>>>
>>> The only difference between the first few datasets and last dataset is:
>>>
>>> All values for the first few data sets > 1.
>>>
>>> The values for the last data sets between 0 and <1.
>>>
>>> Any idea why this happens.
>>>
>>> Your help is highly appreciated.
>>>
>>> Charles
>>>
>>> 
=================================================
==
>>> postscript("Figure.eps", paper="letter", horizontal=FALSE)
>>> par(mfrow=c(3,3))
>>> par(omi=c(2,0.2,1.8,0.2), mai=c(0.4, 0.4, 0.5, 0.1) )
>>>
>>> my.input <- read.table("input.data", header=FALSE, sep="\t")
>>> my.input.exp <- read.table("exp.data", header=FALSE, sep="\t")
>>>
>>> hist(my.input.exp[,2], breaks=40,freq = FALSE,xlab="", border =
>>> "grey30",
>>> ylab="", main="")
>>> hist(my.input[,2], breaks=40,freq = FALSE,xlab="", ylab="", main="",
>>> border
>>> = "red",add =TRUE)
>>> mtext("Fraction", side=2, line=2, cex=0.7)
>>> box()
>>> 
=================================================
=====
>>> --
>>> View this message in context: http://www.nabble.com/Help-needed%3A-
>>> Fraction-for-Histogram-%3E-1-----tp24150345p24150345.html
>>> Sent from the R help mailing list archive at Nabble.com.
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> 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.
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> 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.
>>
>>
>
> --
> View this message in context: http://www.nabble.com/Help-needed%3A-
Fraction-for-Histogram-%3E-1-----tp24150345p24154095.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> 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.
>

Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle