I'm using glmnet for logistic regression, I got a fairly sparse dataset, 20,000 samples(very imbalanced too, 5% from one group), 1500 variables,. the code have beed running for 2 hours, still waiting for result, I am doing lasso here(alpha=1), my computer is core 2 due CPU @3Ghz, 4GB ram, why it's much more slower than the speed report in tibshirani etc's paper? has anyone got same problem? Many thanks yan -- View this message in context: http://r.789695.n4.nabble.com/glmnet-speed-tp4629953.html Sent from the R help mailing list archive at Nabble.com.