[R-sig-Geo] Can I use covariates in addition to spatial coordinates for 'qreglwr' in the McSpatial package?

Joelle k. Akram chino_tones at hotmail.com
Wed Feb 21 22:31:51 CET 2018


Greetings all,


I am trying to run the following function spatial regression:

https://rdrr.io/cran/McSpatial/man/qreglwr.html


Based on the examples given below the weblink, the function is run with spatial coordinates:

fit <- qreglwr<https://rdrr.io/cran/McSpatial/man/qreglwr.html>(LNDENS~LONGITUDE+LATITUDE,taumat=c<https://rdrr.io/r/base/c.html>(.10,.50,.90),window<https://rdrr.io/r/stats/window.html>=.20,data<https://rdrr.io/r/utils/data.html>=cookdata<https://rdrr.io/cran/McSpatial/man/cookdata.html>)


And it is run without spatial coordinates (i.e., only covariates) :

fit1 <- qreglwr<https://rdrr.io/cran/McSpatial/man/qreglwr.html>(LNDENS~DCBD,taumat=c<https://rdrr.io/r/base/c.html>(.10,.50,.90),window<https://rdrr.io/r/stats/window.html>=.30,kern="rect",data<https://rdrr.io/r/utils/data.html>=cookdata<https://rdrr.io/cran/McSpatial/man/cookdata.html>)

In my problem, I am thinking along the lines of GWR where we specify both covariates and spatial coords.

But I do not see it in this function.


Unless I  am missing something, I believe we cannot do the following since the function does may not differentiate between

a spatial coord. variable and a covariate:


fit1 <- qreglwr<https://rdrr.io/cran/McSpatial/man/qreglwr.html>(LNDENS~DCBD+LONGITUDE+LATITUDE,taumat=c<https://rdrr.io/r/base/c.html>(.10,.50,.90),window<https://rdrr.io/r/stats/window.html>=.30,kern="rect",data<https://rdrr.io/r/utils/data.html>=cookdata<https://rdrr.io/cran/McSpatial/man/cookdata.html>)

Any insights are welcome.
cheers.

	[[alternative HTML version deleted]]



More information about the R-sig-Geo mailing list