| constructKnots {Hmsc} | R Documentation |
constructKnots
Description
Construct a Regular Grid of Knot Locations for Spatial GPP Model
Usage
constructKnots(sData, nKnots = NULL, knotDist = NULL, minKnotDist = NULL)
Arguments
sData |
a dataframe containing spatial or temporal coordinates of units of the random level |
nKnots |
the number of knots wanted on the spatial dimension with the shortest range |
knotDist |
the distance between the wanted knots |
minKnotDist |
the minimum distance of a knot to the nearest data point |
Details
This is a helper function for spatial Hmsc models with the
spatial method set to GPP where user must provide knot
locations. Knot locations with a distance greater than
minKnotDist to the nearest data point are dropped from
the grid. If the input locations are
SpatialPoints data, these are treated like
Euclidean coordinates, and if the points are not projected, a
warning is issued.
Only one of nKnots and minKnotDist arguments can be provided.
Value
a data frame with knot locations
Examples
#Creating knots for some 2 dimensional spatial data
n = 100
xycoords = matrix(runif(2*n),ncol=2)
xyKnots = constructKnots(xycoords,knotDist = 0.2, minKnotDist = 0.5)