[R-sig-Geo] Question of ppm warnings of print.ppm

caspar hallmann caspar.hallmann at gmail.com
Mon Sep 27 17:56:58 CEST 2010


Hi Adrian,

you are right, the latest version does not behave like that.
Thanks a lot

Caspar

On Mon, Sep 27, 2010 at 11:57 AM, Adrian Baddeley
<adrian at maths.uwa.edu.au> wrote:
> caspar hallmann <caspar.hallmann at gmail.com> writes:
>
>> I am attempting to fit an inhomogeneous poisson point process to some
>> point pattern data, using ppm from spatstat (v 1.17-5).
>
> This version is out-of-date.
> The current version of spatstat is 1.20-3.
> Try upgrading the package and seeing if you still encounter this problem.
>
> Adrian Baddeley
>
> ====
>
>> I have been found guilty of having NA values in my
>> covariate-images (a categorical variable),
>> and have thus been receiving a lot of warnings.
>
>> Having fitted my model and in attempt to print a summary
>> (or the ppm model) my screen is cluttered with
>
>  > ..$ na.covariates:List of 3
>>   .. ..$ covnames.na  : chr "habitat"
>  >  .. ..$ quadpoints.na: logi [1:4219, 1]
>>              FALSE FALSE FALSE FALSE FALSE FALSE ...
>>  .. ..$ print        : chr "Values of the covariate ?habitat? were NA
>> or undefined at 7.8% (327 out of 4219) of the quadrature points"
>
>> replicated uncountable times, and not allowing me to see any of the
>> useful results.
>
>
>
>



More information about the R-sig-Geo mailing list