Prepare a list of data required for the CAR model; this is for working with (large) raster data files only. For non-raster analysis, see prep_car_data.
prep_car_data2(row = 100, col = 100, quiet = FALSE)Griffith, Daniel A. (2000). Eigenfunction properties and approximations of selected incidence matrices employed in spatial analyses. Linear Algebra and its Applications 321 (1-3): 95-112. doi:10.1016/S0024-3795(00)00031-8 .
Number of rows in the raster
Number of columns in the raster
Controls printing behavior. By default, quiet = FALSE and the range of permissible values for the spatial dependence parameter is printed to the console.
A list containing all of the data elements required by the CAR model in stan_car.
Prepare input data for the CAR model when your dataset consists of observations on a regular (rectangular) tessellation, such as a raster layer or remotely sensed imagery. The rook criteria is used to determine adjacency. This function uses Equation 5 from Griffith (2000) to generate approximate eigenvalues for a row-standardized spatial weights matrix from a P-by-Q dimension regular tessellation.
This function can accommodate very large numbers of observations for use with stan_car; for large N data, it is also recommended to use slim = TRUE or the drop argument. For more details, see: vignette("raster-regression", package = "geostan").
row = 100
col = 120
car_dl <- prep_car_data2(row = row, col = col)