Variance explained with the GSIF package

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Variance explained with the GSIF package

Manuel Spínola
Dear list members,

I am using the fit.gstatModel from the GSIF package.

I obtained 2 different values for variance explained using randomForest.
One is for the model and the other for the prediction.  What is the
difference among them and what is more important to report?

omm <- fit.gstatModel(meuse, om~dist+ffreq, meuse.grid,
method="randomForest")

> omm@regModel

Call:
 randomForest(formula = formulaString, data = rmatrix.s, importance =
TRUE,      na.action = na.omit)
               Type of random forest: regression
                     Number of trees: 500
No. of variables tried at each split: 1

          Mean of squared residuals: 5.952434
                    % Var explained: 49.16


om.rk <- predict(omm, meuse.grid)

> show(om.rk)
  Variable           : om
  Minium value       : 1
  Maximum value      : 17
  Size               : 153
  Total area         : 4964800
  Total area (units) : square-m
  Resolution (x)     : 40
  Resolution (y)     : 40
  Resolution (units) : m
  Vgm model          : Exp
  Nugget (residual)  : 2.78
  Sill (residual)    : 8.36
  Range (residual)   : 6100
  RMSE (validation)  : 1.672
  Var explained      : 76.1%
  Effective bytes    : 1215
  Compression method : gzip

--
*Manuel Spínola, Ph.D.*
Instituto Internacional en Conservación y Manejo de Vida Silvestre
Universidad Nacional
Apartado 1350-3000
Heredia
COSTA RICA
[hidden email] <[hidden email]>
[hidden email]
Teléfono: (506) 8706 - 4662
Personal website: Lobito de río <https://sites.google.com/site/lobitoderio/>
Institutional website: ICOMVIS <http://www.icomvis.una.ac.cr/>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo