[R-sig-Geo] countingweights?

[Rolf Turner]

On 01/11/12 18:22, salma anwar wrote:
>
> Hi, I am trying  to fit a "Area-Interaction" point process model to my data consisting of log-landing locations in a forest. I get the following warning message alongwith the result of the fitting: "1: In countingweights(id, areas) :
>    Some tiles with zero area contain points
> 2: In countingweights(id, areas) : Some weights are zero" Could you please guide me as to what error this message indicates? Thanks!                		 	   		
>

They aren't errors, they are warnings.  *Usually* they are not anything 
to really
worry about.  They arise when the observation window of the pattern has some
peculiarities; either it is a very "irregular" polygon (usually with 
many edges) or
a window of type "mask".

*Sometimes* the warnings can be made to go away by using a finer grid 
--- set
the argument "nd" of ppm() to a large value --- larger than the default; 
e.g.
to 256 or 512.  (This will have the impact of slowing down the calculations
substantially.)  Sometimes refining the pixellation of the mask type window
(or converting a polygonal window to a mask type window, with a fine 
pixellation)
will make the warnings go away.

However, as I said initially, it *usually* doesn't matter.

Note that the fact that you are fitting an area-interaction model is of no
relevance.  You'd get the same warnings fitting any model, including a
Poisson model if you set forcefit=TRUE.

     cheers,

         Rolf Turner