[R-sig-ME] spatial correlation structure for GLS (nlme)

[ONKELINX, Thierry]

Dear Augustin,

Which residuals did you use? You need to normalised residuals because they take the effect of the correlation function into account.

Best regards,

Thierry

----------------------------------------------------------------------------
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek
team Biometrie & Kwaliteitszorg
Gaverstraat 4
9500 Geraardsbergen
Belgium

Research Institute for Nature and Forest
team Biometrics & Quality Assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium

tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
www.inbo.be

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey
  

> -----Oorspronkelijk bericht-----
> Van: r-sig-mixed-models-bounces at r-project.org 
> [mailto:r-sig-mixed-models-bounces at r-project.org] Namens Ing. 
> Agr." Agustín Alesso
> Verzonden: vrijdag 10 december 2010 12:38
> Aan: r-sig-mixed-models at r-project.org
> Onderwerp: [R-sig-ME] spatial correlation structure for GLS (nlme)
> 
> Hello,
> 
> I need some advice to include a spatial correlation structure 
> to an ANOVA one-way GLS model. 
> 
> First, I fit a simple model asuming uncorrelated errors and 
> plotted the residuals variogram which showed the spatial 
> autocorrelation. Then, with gstat package, I fit some models 
> to guess initial values for corSpatial.
> The best model fitted has a nugget efect with to spherical 
> structures (2 ranges).
> Before adding more complexity to the model, I tried to fit 
> few models with basic predefined corStructures and then 
> looking at their residuals variograms. Although including 
> correlation results in smaller AICs, the variograms still 
> showing strong spatial correlation of residuals.  
> 
> Is there a way to bluid a spatial correlation structure 
> including two ranges?
> 
> Thanks in advance,
> 
> 
> 
> --
> Ing. Agr. Agustín Alesso
> Dpto. Ciencias del Ambiente
> Fac. Cs. Agrarias - UNL
> Kreder 2805 - S3080HOF - Esperanza
> Tel: 03496-428575 int 337/256
> 
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>