similar to: A question about lars

Hi here is the code as example lars is in package lars > x<-matrix(rnorm(20*5,0,1),20,5) > bs<-matrix(sample(seq(1:10),5),5,1) > er<-rnorm(20,0,1) > y<-x%*%bs+er > lobj<-lars(x,y,type="lasso") > names(lobj) [1] "call" "type" "df" "lambda" "R2" [6] "RSS"

Hi, I am writing to seek some guidance regarding using Lasso regression with the R package LARS. I have introductory statistics background but I am trying to learn more. Right now I am trying to duplicate the results in a paper for shRNA prediction "An accurate and interpretable model for siRNA efficacy prediction, Jean-Philippe Vert et. al, Bioinformatics" for a Bioinformatics project

Hello, I have a question about lars package, probably basic. The returned values of lars function include R squares along the variable selection path. However, such values are always slightly different from the R squares returned by the regression function lm using the same models. Anyone know the reasons? Very important, and needs quick answers. Thanks a million! -- View this message in

Hi, I have a table with data, as below: dput(table): structure(list(Adrian.Cole = c(0L, 0L, 0L, 0L, 0L, 0L), Alison.Wong = c(0L, 0L, 0L, 0L, 0L, 0L), Andrei.Savu = c(0L, 0L, 0L, 0L, 0L, 0L), Bruno.Dumon = c(0L, 0L, 0L, 0L, 0L, 0L), Edward.J..Yoon = c(0L, 0L, 0L, 0L, 0L, 0L), Eugene.Koontz = c(0L, 0L, 0L, 0L, 0L, 0L), Jakob.Homan = c(0L, 0L, 0L, 0L, 0L, 0L), Kelvin.Kakugawa = c(0L, 0L,

I am trying to extract coefficients from lars fit and can't find how to get intercept. E.g. y = rnorm(10) x = matrix(runif(50),nrow=10) X = data.frame(y,x) fit1 = lars(as.matrix(X[,2:6]),as.matrix(X[,1])) fit2 = lm(y~.,data=X) Then, if I do: > predict(fit1,s=1,mode='fraction',type='coefficients')$coef X1 X2 X3 X4 X5 0.3447570

Hi, all, I am using the lars package for lasso estimate. So I get a lasso fit first: lassofit = lars(x,y,type ="lasso",normalize=T, intercept=T) Then I want to get coefficient with respect to a certain value of \lambda (the tuning parameter), I know lars has three mode options c("step", "fraction", "norm"), but can I use the \lambda value instead

Hi,   I'm using lars package to run some regression analysis and my doubt now is how can I predict my model to another dataset? Let me explain a little better: I have a dataset from which I withhold some data. With the data that wasn't withheld, I create the model. Now, what I'm not being able to do is apply the model back to the data that I withheld. Any suggestions?   Here it goes

Hi, I installed package 'lars' earlier tonight and did not get any sort of error message. ### ** building package indices ... ** testing if installed package can be loaded Loaded lars 0.9-8 * DONE (lars) ### However, when I try to load it, I get ### > library(lars, lib.loc = "~/R/x86_64-redhat-linux-gnu-library/2.11") Error : .onLoad failed in loadNamespace() for

I am trying to use the lars package in R to carry out lasso analysis. However, I am having some problems. Please could you help me with the following questions: 1) Exactly what format do x and y need to be in for cv.lars(x, y) and lars (x, y)? And what information do x and y need to contain exactly? I have tried using to test just a simple matrix of numeric values for x and a simple vector of

We have been using Least Angle Regression (lars) to help identify predictors in models where the outcome is continuous. To do so we have been relying on the lars package. Theoretically, it should be possible to use the lars procedure within a general linear model (glm) framework - we are particular interested in a logistic regression model. Does anyone have examples of using lars with logistic

hello I tried to use lars() but neither with my own data nor with the sample data it works. I get in both cases the following error prompt: > data(diabetes) > par(mfrow=c(2,2)) > attach(diabetes) > x<-lars(x,y) Error in one %*% x : requires numeric matrix/vector arguments > x<-lars(x,y, type="lasso") Error in one %*% x : requires numeric matrix/vector arguments

Hi, I'm having trouble running 'lars'. When I install it I get the following warning: >install.packages('lars') Warning in install.packages("lars") : argument 'lib' is missing: using 'C:\Users\Anna\Documents/R/win-library/2.10' --- Please select a CRAN mirror for use in this session --- trying URL

I'm having the exact problem outlined in a previous post from 2005 - unfortunately the post was never answered: http://tolstoy.newcastle.edu.au/R/help/05/10/15055.html When running: lm2=lars(x2,y,type="lasso",use.Gram=F) I get an error: Error in if (zmin < gamhat) { : missing value where TRUE/FALSE needed ...when running lasso via lars() on a 67x3795 set of predictors. I

dear all, It appears that glmnet(), when "selecting" the covariates entering the model, skips from K covariates, say, to K+2 or K+3. Thus 2 or 3 variables are "added" at the same time and it is not possible to obtain a ranking of the covariates according to their importance in the model. On the other hand lars() "adds" the covariates one at a time. My question

Hi! When I apply the lars (least-angle-regression) method to my data (3655 features, only 355 data points, no I did not mistype), I observe a strange behaviour: 1) The beta values tend to grow into real high values quite fast up to a point where they overflow and get negative. The overflow is not a problem, I don't need the last part of the analysis anyway, but why do they just

When plotting a lars object, I cannot find a way to plot solid lines. Even when the arguments breaks=F and lty="solid" are used, the vertical lines at the break points do not plot but asterisks indicating the breaks still plot as part of each path leaving solid lines broken up by asterisks at the break points. I'm using the following code. larsfit <-

Hello, I have problems with 'lars' package. I found the previous post of the person who had the same issue, but the suggested solution in that post did not help me. I created the matrices: ResponseMatrix <- data.frame (GAOdecision=GAOdecision) PredictorsMatrix <- data.frame (WeaponvsNon = WeaponvsNon, ProductvsService = ProductvsService, KDuration = KDuration, BusinessSize =

Dear all, I'm using zeroinfl() from the pscl-package for zero inflated Poisson regression. I would like to calculate (aproximate) prediction intervals for the fitted values. The package itself does not provide them. Can this be calculated analyticaly? Or do I have to use bootstrap? What I tried until now is to use bootstrap to estimate these intervals. Any comments on the code are welcome.

Hi All, when i run cv.lars(x3,y3) it runs fine. but when i run cv.lars(x3,y3,fraction=seq(0,0.1,100)) I get the following error. Error in apply((y[omit] - fit)^2, 2, mean) : dim(X) must have a positive length Any help/suggestions will be appreciated. Thanks. Harsh --------------------------------- [[alternative HTML version deleted]]

There may be benefits to having a machine learning method that explicitly targets covariate balance. We have experimented with optimizing the weights directly to obtain the best covariate balance, but got some strange solutions for simple cases that made us wary of such methods. Machine learning methods that yield calibrated probability estimates should do well (e.g. those that optimize the