[R] Error in lasso regression (linear dependence issue?)

YaoPau jonrgregg at gmail.com
Thu May 17 10:55:15 CEST 2012


I have a rather large dataset which I just inserted hundreds of interaction
terms, and I guess R is saying there is a linear dependence when I run: 
l1ce(y~., data=x, bound = .5, absolute.t = FALSE).

The error is below:

/Error in if (any(bound < 0)) stop("'bound'(s) must be non negative") : 
  missing value where TRUE/FALSE needed
In addition: Warning message:
In l1ce(poss6 ~ ., data = x, bound = 0.5, absolute.t = FALSE) :
  X Matrix (transformed variables) has rank 1081 < p = 1086, i.e., is
deficient/

Google doesn't show anything regarding the first error.  Any ideas what is
causing this?

Also, how can I find which column vectors have a linear dependence?

--
View this message in context: http://r.789695.n4.nabble.com/Error-in-lasso-regression-linear-dependence-issue-tp4630347.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list