Hi, in R there is the function "density" which computes kernel density estimates. Is there a "cumulative" version of it? Something like they have in Matlab: http://www.mathworks.nl/help/toolbox/stats/ksdensity.html I know there is ecdf, but I'm not sure it's based on kernel density smoothing. Thanks -- View this message in context: http://r.789695.n4.nabble.com/Cumulative-density-kernel-smoothing-tp4384652p4384652.html Sent from the R help mailing list archive at Nabble.com.
Cumulative density (kernel smoothing)
2 messages · francogrex, David Winsemius
On Feb 13, 2012, at 1:01 PM, francogrex wrote:
Hi, in R there is the function "density" which computes kernel density estimates. Is there a "cumulative" version of it? Something like they have in Matlab:
I'm not aware of one, but you could use `integrate`. You will need to limit your range or use the version in package logspline if your range is fixed. You will also need to account for the fact that the inteegral will only be approximately == 1. Searching the Archives will bring up several similar questions from prior years.\
http://www.mathworks.nl/help/toolbox/stats/ksdensity.html I know there is ecdf, but I'm not sure it's based on kernel density smoothing. Thanks
`ecdf` is not based on density.
David Winsemius, MD West Hartford, CT