On Tue, 6 Sep 2011 15:43:45 -0700 (PDT) antie66 <anti...@hotmail.com> wrote:
Hello Louise, > I recently discovered it is possible to conduct home-range analyses > in QGIS using R through Python and I am desperately trying to find a > way to analyse my data (locational data from a baboon troop, > available in shapefile, .csv, .gpx file or whatever file it takes). I > want to analyse my data using a fixed kernel density using a plug-in > method (as proposed by Sheather & Jones (1991) and described in > detail in Section 3.6 of Wand & Jones (1995)). I've tried using the > 'dpik' function in the 'KernSmooth' package in R directly, but I am > too inexperienced too make it work (very frustrating!). The formats accepted by HomeRange plugin are the ones that QGIS can read, so all that you mentioned. Unfortunately the plugin doesn't implement the plug-in method, as it relies mainly on R's adehabitat package. It offers 3 methods for h estimation: h ref (fixed bandwidth), h LSCV (Least Squares Cross Validation) and adjusted h (Wauters et al., 2007). It also proposes NNCH (Nearest-Neighbour Convex Hull, Getz et al., 2007) as an alternative method to overcome some of the issues that affect kernels. If none of these techniques is better than kernel with plug-in method, I'm willing to help you getting it to work in R directly, and afterwards add it to HomeRange plugin. > Can anybody let me know whether this would be possible in QGIS, > before I start the what sounds like a difficult download process? Paolo and Giovanni gave very good pointers, if you need any help just write also to animove list (in cc). > Many thanks for any advice offered. > > Louise Hope this helps, and all the best for your research, Anne -- http://gis.cri.fmach.it/ghisla/ Wauters, L. A., Preatoni, D. G., Molinari, A. e Tosi, G. (2007). Radio tracking squirrels: performance of home range density and linkage estimators with small range and sample size. Ecol. Model., 3-4:333–344. Getz, W. M., Fortmann-Roe, S., Cross, P. C., Lyons, A. J., Ryan, S. J. e Wilmers, C. C. (2007). Locoh: Nonparametric kernel methods for constructing home range and utilization distributions. PLoS ONE, 2.
signature.asc
Description: PGP signature
_______________________________________________ AniMov mailing list AniMov@faunalia.it http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov