search for: adalasso

...ARS algorithm (ref: The Adaptive Lasso and Its Oracle Properties, Zou 2006) for adaptive lasso penalized linear regression. 1. w(j) <- |beta_ols(j)|^(-gamma) gamma>0 and j = 1,...,p 2. define x_new(j) <- x(j)*w(j) 3. apply LARS to solve modified lasso problem out.adalasso <- lars(X_new,y,type="lasso") or enet(X_new, y, lambda=0) suppose, the estimated coefficients are beta_new(j), j=1,...,p 4. beta_adalasso <- beta_new(j)*w(j) everything went fine except I am having problem in plotting standardized (adaptive lasso) coefficients vs. |beta|/max(|be...