[R-sig-Geo] New verison of GWmodel on CRAN

[binbinlu at whu.edu.cn]

Dear all,

As you noticed, there are 2.0-x versions of GWmodel available on CRAN recently.  In cotrast to the 1.x-x versions, there are a number of new features, as follows:
1)  Some of the more computationally intensive functions have been re-coded with C++ by linking to Rcpp and RcppArmadillo. Their computational efficiency has been greatly improved.
2)  In order to make the function names more understandable, some of the funcitons have been re-named:
    montecarlo.gwss	---> gwss.montecarlo
    montecarlo.gwpca.1	--->  gwpca.montecarlo.1
    montecarlo.gwpca.2 	---> gwpca.montecarlo.2
    model.selection.gwr	---> gwr.model.selection
    model.view.gwr	--->  gwr.model.view
    model.sort.gwr	--->  gwr.model.sort
    montecarlo.gwr ---> gwr.montecarlo
    writeGWR 	---> gwr.write
    writeGWR.shp ---> gwr.write.shp
    glyph.plot	 ---> gwpca.glyph.plot
    check.components 	---> gwpca.check.components
    mink.approach ---> gwr.mink.approach
    matrixview 	---> gwr.mink.matrixview
    gwr.generalised ---> ggwr.basic (A reported bug has been fixed with it, and now it is fine with GWmodel_2.0-2 or GWmodel_2.0-3)

Note that we will keep the accesses to to the old names, to ensure the example code work well in the two published vignettes Lu et a. (2014) and Gollini et al. (2015):

    Gollini I, Lu B, Charlton M, Brunsdon C, Harris P (2015) GWmodel: an R Package for exploring Spatial Heterogeneity using Geographically Weighted Models. Journal of Statistical Software, 63(17):1-50, http://www.jstatsoft.org/v63/i17/ 
    Lu B, Harris P, Charlton M, Brunsdon C (2014) The GWmodel R Package: further topics for exploring 17(2): 85-101,http://www.tandfonline.com/doi/abs/10.1080/10095020.2014.917453 

3) New features and functions are included:
   
   bw.gw.average: Select bandwidth for GW averages;
   gwr.mink.pval: Select the values of p for the Minkovski approach for GWR (see details in Lu et al. 2016: Lu, B, Charlton, M, Brunsdon, C & Harris, P(2016). The Minkowski approach for choosing the distance metric in Geographically Weighted Regression. International Journal of Geographical Information Science, 30(2): 351-368. )
   For an adaptive kernel, the bandwidth (the number of nearest neighbours) can be specified as a larger number than the number of observations, and in this case the real value for calculating the weights is {the maximum distance between regression location and all the observations} * (bandwidth/number of observations)

4) Bugs repaired for gwr.basic, gw.dist and ggwr.basic (Please update or install the latest version: GWmodel_2.0-3)

Do feel free to make any suggestions/comments on the new versions. Many thanks.

Best regards,

GWmodel Developement Team