R help - Nov 2011 - GAM

Hi R community!
I am analyzing the data set "motorins" in the package
"faraway" by using
the generalized additive model. it shows the following error. Can some one
suggest me the right way?

library(faraway)
data(motorins)
motori <- motorins[motorins$Zone==1,]
library(mgcv)
>amgam <- gam(log(Payment) ~ offset(log(Insured))+
s(as.numeric(Kilometres)) + s(Bonus) + Make + s(Claims),family = gaussian,
data = motori)
Error in smooth.construct.tp.smooth.
spec(object, dk$data, dk$knots) :
  A term has fewer unique covariate combinations than specified maximum
degrees of freedom
> summary(amgam)
Error in summary(amgam) : object 'amgam' not found
Gyan

	[[alternative HTML version deleted]]

Gyanendra Pokharel wrote on 11/08/2011 10:57:38 AM:
> 
> Hi R community!
> I am analyzing the data set "motorins" in the package
"faraway" by using
> the generalized additive model. it shows the following error. Can some 
one
> suggest me the right way?
> 
> library(faraway)
> data(motorins)
> motori <- motorins[motorins$Zone==1,]
> library(mgcv)
> >amgam <- gam(log(Payment) ~ offset(log(Insured))+
> s(as.numeric(Kilometres)) + s(Bonus) + Make + s(Claims),family = 
gaussian,
> data = motori)
> Error in smooth.construct.tp.smooth.
> spec(object, dk$data, dk$knots) :
>   A term has fewer unique covariate combinations than specified maximum
> degrees of freedom
> > summary(amgam)
> Error in summary(amgam) : object 'amgam' not found
> Gyan

You could try reducing the number of degrees of freedom in the smooth 
terms with the k argument of the s() function:

amgam <- gam(log(Payment) ~ offset(log(Insured))+ 
s(as.numeric(Kilometres), k=3) + 
        s(Bonus, k=3) + Make + s(Claims, k=3), data=motori) 

Jean
	[[alternative HTML version deleted]]

Kilometres has only 5 unique values, while Bonus has only 7, but the 
default smoothing basis dimension for the s terms is 10, so there is a 
problem. Solution is to reduce the basis dimension. e.g.

amgam <- gam(log(Payment) ~ offset(log(Insured))+
+ s(as.numeric(Kilometres),k=5) + s(Bonus,k=7) + Make + s(Claims),family 
= gaussian,
+ data = motori)


On 08/11/11 16:57, Gyanendra Pokharel wrote:
> Hi R community!
> I am analyzing the data set "motorins" in the package
"faraway" by using
> the generalized additive model. it shows the following error. Can some one
> suggest me the right way?
>
> library(faraway)
> data(motorins)
> motori<- motorins[motorins$Zone==1,]
> library(mgcv)
>> amgam<- gam(log(Payment) ~ offset(log(Insured))+
> s(as.numeric(Kilometres)) + s(Bonus) + Make + s(Claims),family = gaussian,
> data = motori)
> Error in smooth.construct.tp.smooth.
> spec(object, dk$data, dk$knots) :
>    A term has fewer unique covariate combinations than specified maximum
> degrees of freedom
>> summary(amgam)
> Error in summary(amgam) : object 'amgam' not found
> Gyan
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>

-- 
Simon Wood, Mathematical Science, University of Bath BA2 7AY UK
+44 (0)1225 386603               http://people.bath.ac.uk/sw283

On 11/08/2011 11:57 AM, Gyanendra Pokharel wrote:
> Hi R community!
> I am analyzing the data set "motorins" in the package
"faraway" by using
> the generalized additive model. it shows the following error. Can some one
> suggest me the right way?
>
> library(faraway)
> data(motorins)
> motori<- motorins[motorins$Zone==1,]
> library(mgcv)
> amgam<- gam(log(Payment) ~ offset(log(Insured))+
> s(as.numeric(Kilometres)) + s(Bonus) + Make + s(Claims),family = gaussian,
> data = motori)
> Error in smooth.construct.tp.smooth.
> spec(object, dk$data, dk$knots) :
>    A term has fewer unique covariate combinations than specified maximum
> degrees of freedom
Just to provide further information,

 > table(motori$Kilometres)

  1  2  3  4  5
61 63 60 57 54

Kilometres only has 5 unique values; this complicates the issue of 
estimating a curve, as you only have 5 points to work with.  Likewise, 
Bonus only has 7 unique values.  Claims does not pose any problem, as it 
has 98 unique values.  As Jean suggested, you can get around this issue 
with:

amgam<- gam(log(Payment) ~ offset(log(Insured)) + 
s(as.numeric(Kilometres),k=5) + s(Bonus,k=7) + Make + s(Claims),family = 
gaussian, data = motori)

-- 
Patrick Breheny
Assistant Professor
Department of Biostatistics
Department of Statistics
University of Kentucky