Aggregation problems

I would like to extend this question. Is it possible to vary the
size of the step, this 'window' is passed over the area so that the overlap
can be chosen. Furthermore it would be usefull to calculate not only
the mean but also other statistical values in the aggregated area.
I'm not a very
experienced GRASS user but i wonder if r.mapcalc can do that job.

        ____ Stefan Jaeger
       / __ Geographisches Institut
       \ | / Universitaet Heidelberg
         \ ____|__ ___|_ INF 348
          ) | ____ | ___ _ W-6900 Heidelberg
         / | (____) | ___\ |/ | Germany
_____ / (___ \____ | |___| | |

                              BITNET: ca4@dhdurz1
                              INTERNET: ca4@vm.urz.uni-heidelberg.de
                              OR: ca4@129.206.98.101