R help - Aug 2009 - compare observed and fitted GAM values

Hi,

I am comparing the observed and fitted values of my GAM model, which includes
the explanatory variables: longitude, depth, ssh, year and month. When I compare
observed and fitted values for longitude, depth and ssh it works. But when I try
to do it for month and year (which are as factors in the GAM model) it
doesn't work. My observed and fitted values are exactly the same.. How is
that possible?  Thanks
> Obs_factor1<-aggregate(x=albdata$turtles,by=list(albdata$Year),FUN=mean)
> names(Obs_factor1)=c("Bin","Observed")
> Obs_factor1
    Bin    Observed
1  1997 0.017094017
2  1998 0.010652463
3  1999 0.023000000
4  2000 0.017167382
5  2001 0.030465950
6  2002 0.007446809
7  2003 0.010568032
8  2004 0.011450382
9  2005 0.016270338
10 2006 0.017006803
11 2007 0.030969031
12 2008 0.066455696
>
Fit_factor1<-aggregate(x=predict(gam.def.lon,type="response"),by=list(albdata$Year),FUN=mean)
> names(Fit_factor1)=c("Bin","Fitted")
> Fit_factor1
    Bin      Fitted
1  1997 0.017094017
2  1998 0.010652463
3  1999 0.023000000
4  2000 0.017167382
5  2001 0.030465950
6  2002 0.007446809
7  2003 0.010568032
8  2004 0.011450382
9  2005 0.016270338
10 2006 0.017006803
11 2007 0.030969031
12 2008 0.066455696





      
	[[alternative HTML version deleted]]

On Aug 21, 2009, at 5:03 PM, Luc?a Rueda wrote:
> Hi,
>
> I am comparing the observed and fitted values of my GAM model, which  
> includes the explanatory variables: longitude, depth, ssh, year and  
> month. When I compare observed and fitted values for longitude,  
> depth and ssh it works. But when I try to do it for month and year  
> (which are as factors in the GAM model) it doesn't work. My observed  
> and fitted values are exactly the same..
So then,  you are complaining because it "works" but _unexpectedly_  
well.
> How is that possible?  Thanks
If you have a balanced design in Year and it is the only covariate,  
then that is precisely what should happen with any regression method:

 > newdat <- data.frame(val=rnorm(40), cat=factor(1:4))
 > aggregate(newdat$val, list(newdat$cat), mean)
   Group.1          x
1       1  0.4972092
2       2 -0.1042936
3       3  0.1549305
4       4 -0.1500513
 > lm(val~cat, newdat)

Call:
lm(formula = val ~ cat, data = newdat)

Coefficients:
(Intercept)         cat2         cat3         cat4
      0.4972      -0.6015      -0.3423      -0.6473

 > 0.4972092 + coef(lm(val~cat, newdat))[2:4]
       cat2       cat3       cat4
-0.1042937  0.1549304 -0.1500514
> snipped output from OP

-- 
David Winsemius, MD
Heritage Laboratories
West Hartford, CT

You haven't given quite enough information to be sure, but I would guess
that
this is not really a problem, but rather the interesting proporty of GLMs 
fitted with a canonical link described in e.g. section 2.1.8 of Wood (2006) 
Generalized additive models: and introduction with R, or at the beginning of  
the GLM chapter in Venables and Ripley MASS. 

On Friday 21 August 2009 22:03, Luc??a Rueda wrote:
> Hi,
>
> I am comparing the observed and fitted values of my GAM model, which
> includes the explanatory variables: longitude, depth, ssh, year and month.
> When I compare observed and fitted values for longitude, depth and ssh it
> works. But when I try to do it for month and year (which are as factors in
> the GAM model) it doesn't work. My observed and fitted values are
exactly
> the same.. How is that possible?? Thanks
>
> >
Obs_factor1<-aggregate(x=albdata$turtles,by=list(albdata$Year),FUN=mean)
> > names(Obs_factor1)=c("Bin","Observed")
> > Obs_factor1
>
> ??? Bin??? Observed
> 1? 1997 0.017094017
> 2? 1998 0.010652463
> 3? 1999 0.023000000
> 4? 2000 0.017167382
> 5? 2001 0.030465950
> 6? 2002 0.007446809
> 7? 2003 0.010568032
> 8? 2004 0.011450382
> 9? 2005 0.016270338
> 10 2006 0.017006803
> 11 2007 0.030969031
> 12 2008 0.066455696
>
> >
Fit_factor1<-aggregate(x=predict(gam.def.lon,type="response"),by=list(alb
> >data$Year),FUN=mean)
names(Fit_factor1)=c("Bin","Fitted")
> > Fit_factor1
>
> ??? Bin????? Fitted
> 1? 1997 0.017094017
> 2? 1998 0.010652463
> 3? 1999 0.023000000
> 4? 2000 0.017167382
> 5? 2001 0.030465950
> 6? 2002 0.007446809
> 7? 2003 0.010568032
> 8? 2004 0.011450382
> 9? 2005 0.016270338
> 10 2006 0.017006803
> 11 2007 0.030969031
> 12 2008 0.066455696
>
>
>
>
>
>
> 	[[alternative HTML version deleted]]
-- 
> Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK
> +44 1225 386603  www.maths.bath.ac.uk/~sw283