Interpretation of Variogram for model selection

[jillsh...@gmail.com]

Dear Chris

 

I have been working my way through your Variogram and Model Selection vignette using the cilla data.

 

I have eventually ended up at the Model Selection using Maximum Likelihood section and run the following code:

 

FITZ <- ctmm.select(cilla, m2, verbose = TRUE, level = 1)

summary(FITZ)

 

                    dAICc DOF[mean]

OUF anisotropic    0.0000 13.216116

OUF isotropic    113.9431 13.876278

OU anisotropic  1458.2972  6.648584

 

#Use a variogram to check that the ML process has selected the best model

par(mfcol=1:2)

plot(SVF,CTMM=FITZ,col.CTMM=c("red","purple","blue"),fraction=0.65,level=0.5)

title("zoomed out")

plot(SVF,CTMM=FITZ,col.CTMM=c("red","purple","blue"),fraction=0.002,level=0.5)

title("zoomed in")

 

this produced this figure:

 

It would seem that the best model is the one described by the red line. This should correspond to the OUF anisotropic model. But I am a little confused given that the colour coding has previously represented MO=red, m1=purple and m2=blue.

 

I would just like to check that I understand things correctly. It would seem that In this case the code has ranked the colours according to best AIC score so:

Red = OUF anisotropic

Purple  = OUF isotropic

Blue = OU anisotropic.

 

Many thanks in advance for your assistance.

Jill

 

[Christen Fleming]

Hi Jill,

You have 3 models in your list of models FITZ and 3 colors in your array of colors c("red","purple","blue"), and the colors are matched one-to-one in order. Here it is by increasing AIC, but that is because ctmm.select returns a sorted list of models. If you apply rev() to the model list or color array argument, then the colors will run in reverse. The plot method doesn't do any extra sorting.

Best,
Chris

[Jill Shephard]

Thanks Chris for the quick response. So my interpretation is correct...?

[Christen Fleming]

Yes, but plot didn't sort the models by AIC, ctmm.select did.

Best,
Chris

[Jill Shephard]

Yes, great that was my understanding.

Thanks again

Jill