[R-sig-Geo] model fitting of randomly generated data in spatstat

[Rolf Turner]

On 03/04/15 01:12, robleaf wrote:
> Rolf,
>
> Really appreciate your responding to my query. We were interested in
> simulating highly clumped distributions and the values that I used
>
> rMatClust(kappa = 2, r = 2, mu = 2000))
>
> Seemed to do a adequate job in creating them. I am appreciative of your
> comment to the effect that the parameter values cannot be too strange. Looks
> like I will need to spend more time with this. Really appreciate your in
> input, regards, Robert


But those parameter values *won't* give you a "highly clumped" 
distribution.   You will generate on average *2* clumps and since the 
radius of each "clump" is 2 and you are simulating in the unit square, 
each clump will pretty much cover the window.  So you are (partially) 
superimposing (on average) two *very* dense clouds of points on the unit 
square.  This will not look ***at all*** "clumped.

Have you *plotted* the results of any of your simulations?

cheers,

Rolf

-- 
Rolf Turner
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276
Home phone: +64-9-480-4619