|
3D KDE: msg#00047lang.r.geo
Hi List, I'm looking for help with finding a way to create kernel density surfaces with point count data. The method needs to use the edge correction, so initially I thought the methods in splancs would work. But it looks like kernel2d corrects for edge effects (it has an argument for a study area polygon), but kernel3d does not appear to apply the edge correction. Does anyone know if there is a function that does 3D kernel density with edge correction? Kernel3d has arguments for the following. I would expect that xgr and ygr would be equivalent to the points in the first points? I think I'm probably missing something here. pts: A matrix of event coodinates x,y. times: A vector of event times, t. xgr: The values of x at which to compute the kernel function. ygr: The values of y at which to compute the kernel function. zgr: The values of time at which to compute the kernel function. hxy: The quartic kernel width in the x and y direction. hz: The quartic kernel width in the temporal direction. Any help or suggestions much appreciated, Colin Colin Robertson MSc Candidate Dept of Geography University of Victoria ph. (250)885-8065 [[alternative HTML version deleted]]
|
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | GSTAT fit.variogram method, Zev Ross |
|---|---|
| Next by Date: | Re: GSTAT fit.variogram method, Edzer J. Pebesma |
| Previous by Thread: | GSTAT fit.variogram method, Zev Ross |
| Next by Thread: | GSTAT, karl . sommer |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |