lara harrup (IAH-P)
2009-Apr-29 17:39 UTC
[R] Installing/using "glars" package --- Error in library(glars) : 'glars' is not a valid installed package
Hi all I seem to have fallen at the first hurdle with my analysis, I have a set of binary disease outbreak data linked to a large number of landscape metrics variables and environmental variables which I would like to as predictor variables in a Least Angle Logistic Regression using the glars.fit.s function in the glars package (my data exhibits some multicollinearity hence the LARS) but when I try and load this in R (version 2.9.0) I get the following message> library(glars)Error in library(glars) : 'glars' is not a valid installed package I have the windows binary for the glars package downloaded from the site below [the glars does not seem to be on the CRAN site, (LARS for linear least angle regression is and loads fine)] and unzipped in my library folder. http://csan.insightful.com/PackageDetails.aspx?Package=glars The package is written in the S-dialect but as far as I can tell from the literature it should run in both R and S-PLUS, the description file states: 'Package: glars Title: Generalized Least Angle Regression Version: 0.1.2' 'This is "S+GLARS" for S-PLUS or R' 'R (>= 2.0)' 'Dialect: S-PLUS' Not sure if this is a general problem with loading an S dialect package and I need to add some thing else before loading the library if it is in S rather than R? or if it is a problem with the 'glar' package. Any suggestions/tips most appreciated, thanks in advance. Lara lara.harrup at bbsrc.ac.uk
Seemingly Similar Threads
- randomForest - what is a 'good' pseudo r-squared?
- Error with regsubset in leaps package - vcov and all.best option (plus calculating VIFs for subsets)
- 95% Confidence Intervals for AUC - $auc.samples from the Daim Package
- ANCOVA/glm missing/ignored interaction combinations
- Random Forest Variable Importance Interpretation