side effect of l1fit

["Weijie Cai" <wjcai2001@hotmail.com>]

Dear list,
My program contains a function with l1fit (minimum absolute residual 
regression) to fit a model. However, the side effect of l1fit is, sometimes 
there is no unique solution to it. Now I don't want my program to report 
such warning and let it continue to next step with the current solution, 
what should I do?
eg.
fit<-function(x,y)
{
f<-l1fit(x,y,intercept=T)
a<-f$coefficient[1]
b<-f$coefficient[2]
y<-a^2+b^2
y
}

Thanks!
Weijie

_________________________________________________________________
Tired of spam? Get advanced junk mail protection with MSN 8. 
http://join.msn.com/?page=features/junkmail