licb: L1-Norm based Improved Correlation-based Penalty
Description
Object of the penalty class to handle the L1-Norm based Improved Correlation-Based (LICB) Penalty (Ulbricht, 2010).
Usage
licb (lambda = NULL, ...)
Arguments
lambda
two-dimensional tuning parameter parameter. The first component corresponds to the regularization parameter $\lambda_1$ for the lasso penalty
term, the second one $\lambda_2$ for the $L_1$-norm based correlation penalty term. Both parameters must be nonne
...
further arguments
Value
An object of the class penalty. This is a list with elements
penaltycharacter: the penalty name.
lambdadouble: the (nonnegative) regularization parameter.
first.derivativefunction: This returns the J-dimensional vector of the first derivative of the J penalty terms with
respect to $|\mathbf{a}^\top_j\boldsymbol{\beta|}$.
a.coefsfunction: This returns the p-dimensional coefficient vector $\mathbf{a}_j$ of the J penalty terms.
Details
The improved correlation-based (LICB) penalty is defined as
$$P_{\lambda}^{licb}(\boldsymbol{\beta}) = \lambda_1 \sum_{i=1}^p |\beta_i| + \lambda_2 \sum_{i=1}^{p-1} \sum_{j > i}
\left{\frac{|\beta_i - \beta_j|}{1 - \varrho_{ij}} + \frac{|\beta_i + \beta_j|}{1 + \varrho_{ij}}\right}.$$
The LICB has been introduced to overcome the major drawback of the correlation based-penalized estimator, that is its lack of sparsity.
See Ulbricht (2010) for details.
References
Ulbricht, Jan (2010) Variable Selection in Generalized Linear Models. Ph.D. Thesis. LMU Munich.