Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

I have a dataset with between 10,000 and 100,000 feature values. The number of datapoints is between 1,000 and 10,000. I want to perform a LASSO on this dataset but can't really find any good software to do so. Does anyone have any suggestions?

share|improve this question
    
There is a package in R called lars that handles common procedures related to least angle regression and lasso. cran.r-project.org/web/packages/lars/index.html –  Christopher Aden Apr 25 '11 at 20:19
2  
What have you tried and why hasn't it satisfied you? –  Gael Varoquaux Apr 25 '11 at 20:46
1  
Have you looked at the [lasso] tag? There are several relevant questions there. –  cardinal Apr 26 '11 at 0:54
add comment

1 Answer

Check the article by Wu Chen Hastie Sobel Lange - Genome-wide association analysis by lasso penalized logistic regression - 2009. They mention a 'swindle' that is not hard to implement + then you can simply work with glmnet (there is a new version out recently which promises a performance improvement but I haven't had a chance to check it out).

share|improve this answer
    
I know it's an ancient post, but I didn't understand the swindle they used. Could you elaborate? –  Piotr Sokol Apr 26 '13 at 15:04
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.