Solution Paths for Linear and Logistic Regression Models with Categorical Predictors, with SCOPE Penalty
Create a design matrix of categorical variables with correlated column...
Computes SCOPE logistic predictions
Computes SCOPE predictions
Computes solution for SCOPE logistic models
Compute solution for SCOPE linear models.
Create a design matrix of categorical variables.
Computes solutions for linear and logistic regression models with potentially high-dimensional categorical predictors. This is done by applying a nonconvex penalty (SCOPE) and computing solutions in an efficient path-wise fashion. The scaling of the solution paths is selected automatically. Includes functionality for selecting tuning parameter lambda by k-fold cross-validation and early termination based on information criteria. Solutions are computed by cyclical block-coordinate descent, iterating an innovative dynamic programming algorithm to compute exact solutions for each block.