[R] how to visualise the smoothers of a model averaged GAM ?

[Youssef]

Hi,
If i understand what you mean, you can do 
Plot(M.gam.auto)
You Will get the centered predictions for smooths, you can also extract predictions for each term using (predict(M.gam.auto, type="terms")).

The results you got are thé coefficients for each of you smooths pieces (as k=4, you get 3), but they re not interpretable as they are.

Hope this could help

Youssef

Le 2012-09-05 à 11:50, fmoreira <fmoreira at isa.utl.pt> a écrit :

> Hi,
> 
> It is not clear to me what a model-averaged (sensu MuMIn package) GAM is...
> How is the smoother averaged and how can I see it ?
> 
> My model is:
> 
> M.gam.auto <- gam (Leksite ~ s(Altmean,k=4) + s(Class_unfav_prop, k=4) + 
>               + s(Visib2000, k=4) + s(Ddisturbance, k=4) + s(Steppe_prop,
> k=4)
>                  + s(autocov2, k=4), family=binomial, data= leks)
> 
> When i extract the confidence intervals for the average models (all models
> with delta AIC <2), I get:
> 
>> confint (avgmod)
>                             2.5 %       97.5 %
> (Intercept)           -0.705495919  0.090974395
> s(Altmean).1          -1.032000478  0.846255206
> s(Altmean).2          -0.131526964  3.181165030
> s(Altmean).3          -0.479038766  1.511470593
> s(autocov2).1         -0.594732400  5.030680753
> s(autocov2).2         -3.932830030 18.377288299
> s(autocov2).3         -5.859925892  3.269061133
> s(Ddisturbance).1     -0.324700903  0.609412974
> s(Ddisturbance).2     -0.016172651  5.880936668
> s(Ddisturbance).3     -0.881357264  0.862160976
> s(Steppe_prop).1      -0.002864215  0.002861629
> s(Steppe_prop).2      -0.022374707  0.022389396
> s(Steppe_prop).3       0.191842041  1.242861265
> s(Class_unfav_prop).1 -0.119628517  0.109562147
> s(Class_unfav_prop).2 -0.246237564  0.274136388
> s(Class_unfav_prop).3 -0.588583036  0.189353845
> s(Visib2000).1        -0.003592773  0.003593891
> s(Visib2000).2        -0.012680896  0.012686619
> s(Visib2000).3        -0.489773942  0.206553913
> 
> Not sure of this means nor how to visualize the smoothers...
> Help would be appreciated.
> 
> 
> Francisco Moreira
> Institute of Agronomy
> Technical University of Lisbon
> 
> 
> 
> --
> View this message in context: http://r.789695.n4.nabble.com/how-to-visualise-the-smoothers-of-a-model-averaged-GAM-tp4642302.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.