If you want to count the local maxima in the n x n matrix returned by
kde2d, AND you know there are no ties, you could do something like the
following:
> set.seed(1)
> x <- matrix(sample(10, 25, rep=TRUE), 5, 5)
> x
[,1] [,2] [,3] [,4] [,5]
[1,] 3 9 3 5 10
[2,] 4 10 2 8 3
[3,] 6 7 7 10 7
[4,] 10 7 4 4 2
[5,] 3 1 8 8 3
> sum(x > cbind(0, x[,-5]) & x > cbind(x[,-1], 0) & x >
rbind(x[-1,],
0) & x > rbind(0, x[-5,]))
[1] 4
>
Just be careful that your counting formula matches your definition of
"neighbor" (the above formula does not include diagonal neighbors).
And of course, ties make things more complicated (note that the above
simple algorithm misses the local maximum consisting of two 8's in the
last row.)
-- Tony Plate
Patrick Wang wrote:> Hi,
>
> Does anyone know how to count the number of modes in 2 dimensions using
> kde2d function?
>
> Thanks
> Pat
>
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
>