Skip to content
Prev 257098 / 398506 Next

spatstat regression troubles

On 17/04/11 02:17, Gregory Ryslik wrote:
500?  You said 944 previously.  Doesn't really matter, but.
To make use of a covariate you ***really*** need to have the values of 
the covariate
available at ***all*** points of the observation window.  In your 
situation I think that
the best that you can do is to interpolate between the actual observations.

You could use, I think, the interp() function from the "akima" package.  
Here's a toy
demo:
The interpolated image is a bit rough compared with the truth (which we 
know in this
artificial case) but no worse than what one might reasonably expect.

HTH

         cheers,

             Rolf Turner