Maybe Joren means that the y axis has values greater than 1? If that
is the case, that is certainly not evidence of any problem; the
density can have values larger than 1 and still integrate to 1. (And,
just as a silly example, try "dnorm(0, mean = 0, sd = 0.1)").
Best,
R.
On Nov 15, 2007 11:0
I am not sure what you mean when you say it does not integrate to 1.
Here are a couple of cases, and it seems fine to me:
> x <- density(1:30)
> str(x)
List of 7
$ x: num [1:512] -11.0 -10.9 -10.8 -10.7 -10.6 ...
$ y: num [1:512] 6.66e-05 7.22e-05 7.84e-05 8.49e-05 9.20e-05 ...
Hello,
I have a data set of about 300.000 measurements made by an STM which should
apporximately fix a normal (Gaussian) distribution.
I have imported the data in R and used plot(density()) to get a nice plot of
the distribution which in fact looks like a real Gaussian.
However, the integral over
To run my data in another program my data cannot exceed a kurtosis of 0.8. I'm
wondering if there is a package that can determine if the kurtosis for a trait
is equal to or greater than 0.8 and then determine the appropriate normalizing
methods to reduce the kurtosis to less than 0.8. I would als
4 matches
Mail list logo
