[R] KDE routines for data that is aggregated

[David L Carlson]

If the x and y values are regularly spaced, you could use contour() or persp() to plot the densities. If they are not, you can use density(), loess(), gam(), kriging another function to estimate a smooth surface for the values and then estimate the values over a regular grid and then plot with contour, etc.

-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Saptarshi Guha
Sent: Monday, September 8, 2014 6:57 PM
To: R-help at r-project.org
Subject: [R] KDE routines for data that is aggregated

Hello,
Couldn't think of a better subject line. Rather than a matrix like

x,y
..,..
.,..

I have a matrix like
x,y,n,
..,..,..,
..,..,..

and so on. Also, sum(n) is roughly few hundred million. The number of rows
is <1MM

Are they routines to fit a 2d kde estimate to data provided in this form?
I can sample from the data according to weights given by 'n' but i am
curious if there is something that can use all the data when given a
structure of this form.

Regards
Saptarshi

	[[alternative HTML version deleted]]

______________________________________________
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.