For some datasets, for example when the design matrix is not of full rank, 'glmnet' may have very long run times when fitting the relaxed lasso model, in particular when fitting a Cox based model, making it difficult to get solutions either from glmnet() or cv.glmnet(). In this package, 'glmnetr', we provide a workaround and solve for the non penalized relaxed model where gamma=0 for model structures analogue to 'R' functions like glm() or coxph() of the 'survival' package. If you are not fitting relaxed lasso models, or if you are able to get convergence using 'glmnet', then this package may not be of much benefit to you. Note, while this package may allow one to fit relaxed lasso models that have difficulties converging using 'glmnet', this package does not afford the full function and versatility of 'glmnet'. In addition to fitting the relaxed lasso model this package also includes the function cv.glmnetr() to perform a cross validation to identify hyper-parameters for a lasso fit, much like the cv.glmnet() function of the 'glmnet' package. Additionally, the package includes the function nested.glmnetr() to perform a nested cross validation to assess the fit of a cross validated derived lasso model fit. If though you are fitting not a relaxed lasso model but an elastic-net model, then the R-packages 'nestedcv' , 'glmnetSE' or others may provide greater functionality when performing a nested CV. As with the 'glmnet' package, this package passes most relevant output to the output object and tabular and graphical summaries can be generated using the summary and plot functions. Use of the 'glmnetr' has many similarities to the 'glmnet' package and it is recommended that the user of 'glmnetr' first become familiar with the 'glmnet' package , with the "An Introduction to 'glmnet'" and "The Relaxed Lasso" being especially helpful in this regard.
||Mayo Foundation for Medical Education and Research