Skip to content

Shift the normal curve to the top or near to the top of the histogram

4 messages · AbouEl-Makarim Aboueissa, Jim Lemon, Rui Barradas

#
Dear All:

One more thing.

I want to add the normal curve to the histogram. Is there away to stretch
the peak of the curve to the top of the histogram or at least near to the
top of the histogram.

Please see the code below.


Lizard.tail.lengths <- c(6.2, 6.6, 7.1, 7.4, 7.6, 7.9, 8, 8.3, 8.4, 8.5,
8.6,8.8, 8.8, 9.1, 9.2, 9.4, 9.4, 9.7, 9.9, 10.2, 10.4, 10.8,11.3, 11.9)

x<-seq(5,12, 0.001)

hist(Lizard.tail.lengths, main = "Normal Probability Plot of Lizard Tail
Lengths")

curve(dnorm(x ,mean=mean(Lizard.tail.lengths),sd=sd(Lizard.tail.lengths)),
add=TRUE, col=2, lwd = 2)



with many thanks
abou
______________________
AbouEl-Makarim Aboueissa, PhD
Professor of Statistics
Department of Mathematics and Statistics
University of Southern Maine
#
Hi Abou,
Try this:

library(plotrix)
 curve(rescale(dnorm(x
,mean=mean(Lizard.tail.lengths),sd=sd(Lizard.tail.lengths)),
c(0,6)),add=TRUE, col=2, lwd = 2)

Jim


On Mon, Sep 25, 2017 at 9:35 AM, AbouEl-Makarim Aboueissa
<abouelmakarim1962 at gmail.com> wrote:
#
Dear Jim:

Thank you very much

abou
On Sun, Sep 24, 2017 at 8:18 PM, Jim Lemon <drjimlemon at gmail.com> wrote:

            

  
    
#
Hello,

Try using hist argument 'prob = TRUE' or, which is equivalent, 'freq = 
FALSE'.

hist(..., prob = TRUE)   # or hist(..., freq = FALSE)

This is because like this you will have a density, comparable to a 
parametric density. Note that the peak of the normal will be outside the 
plot area so you will have to adjust the plot area dimensions. In this 
case I've set 'ylim = c(0, 0.35)'.


hist(Lizard.tail.lengths, main = "Normal Probability Plot of Lizard Tail
Lengths", ylim = c(0, 0.35), prob = TRUE)

curve(dnorm(x ,mean=mean(Lizard.tail.lengths),sd=sd(Lizard.tail.lengths)),
add=TRUE, col=2, lwd = 2)


Hope this helps,

Rui Barradas

Em 25-09-2017 00:35, AbouEl-Makarim Aboueissa escreveu: