R help - May 2005 - problem to get a correct output

Dear,

I am new user of 'R' (package e1071) and the first
thing I did was reading the manuals. Although it was
not clear for me how it works.
My problem is the following:
I have a data set of customers data of insurances with
variables  Class, Group Distance Claim. The file
called 'xx.dat'. Using the manuals I think I had
entered the data in the program. I used several
function, but I never get an output.
In fact, I want to use the SVM with Vapnik's loss
function for classification for the response variable
'Claim' and discrete explanatory variables Class,
Group, Distance whitout modelling interaction terms
manually. I want to use epsilon = 0.1, cast penalty
constant (C) cost =1, and  a Gaussian radial basis
function kernel with y =1, i.e. k(x,x')exp(-y||x-x'||伈).
I would like to aply here the classification rule to
the whole data set and construct a 2-by-2 table of
claim and predicted levels of Claim.

I hope you can help me further !!
Many thanks in advance !

best regards,
Dominique

You gave no detail for anyone to provide useful help.  What exact commands
did you use, and what, if any, error/warning did you get?  Please (re-)read
the posting guide.  You may also want to do

  vignette("svmdoc")

Andy
> From: Dominique Knockaert
> 
> Dear,
> 
> I am new user of 'R' (package e1071) and the first
> thing I did was reading the manuals. Although it was
> not clear for me how it works.
> My problem is the following:
> I have a data set of customers data of insurances with
> variables  Class, Group Distance Claim. The file
> called 'xx.dat'. Using the manuals I think I had
> entered the data in the program. I used several
> function, but I never get an output.
> In fact, I want to use the SVM with Vapnik's loss
> function for classification for the response variable
> 'Claim' and discrete explanatory variables Class,
> Group, Distance whitout modelling interaction terms
> manually. I want to use epsilon = 0.1, cast penalty
> constant (C) cost =1, and  a Gaussian radial basis
> function kernel with y =1, i.e. k(x,x')> exp(-y||x-x'||??).
> I would like to aply here the classification rule to
> the whole data set and construct a 2-by-2 table of
> claim and predicted levels of Claim.
> 
> I hope you can help me further !!
> Many thanks in advance !
> 
> best regards,
> Dominique
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! 
> http://www.R-project.org/posting-guide.html
> 
> 
>