Dear Tom,
Actually, R complains about duplicate points. I'm pretty sure some
restaurants are really close to each other, or perhaps in the same building.
Could be some geocoding errors as well. If I were to 'wiggle' the data,
would this be dependent on the resolution of the dataset? I'm still figuring
out the spatial resolution by the way. The dataset is restricted to
restaurants in the Netherlands, however.
Tom, thanks very much for the tip. I was struggling with unique(ppp). But
wiggling seems a better approach, certainly since I don't want to loose any
data.
2009/2/17 Tom Petersen <t...@infra.kth.se>
> Michael,
>
> Although your points might seem cluttered (are there reasons to believe
> that
> they shouldn't be?), you should still check what your spatial resolution
> is,
> and if there are several restaurants occupying the same coordinates - in
> that
> case, only the first one will be considered on import. I experienced that
> in
> spatstat with my dataset with a resolution of 250 m.
>
> If this is the case, one solution is to "wiggle" the data by adding uniform
> random disturbances to both coordinates, with amplitude = half the
> resolution
> (e.g. in my case a uniform distribution over [-125,125]).
>
> /Tom
> ============================
> Tom Petersen
> Transport- och lokaliseringsanalys
> Skolan för arkitektur och samhällsbyggnad
> Teknikringen 78 B
> KTH, 100 44 Stockholm
> Tfn 08-790 68 33, 070-424 00 75
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
