quadrupen-class {quadrupen}

Class "quadrupen"

Description

Class of object returned by any fitting function of the quadrupen package (elastic.net or bounded.reg).

Slots

coefficients:

Matrix (class "dgCMatrix") of coefficients with respect to the original input. The number of rows corresponds the length of lambda1.

active.set:

Matrix (class "dgCMatrix", generally sparse) indicating the 'active' variables, in the sense that they activate the constraints. For the elastic.net, it corresponds to the nonzero variables; for the bounded.reg function, it is the set of variables reaching the boundary along the path of solutions.

intercept:

logical; indicates if an intercept has been included to the model.

mu:

A vector (class "numeric") containing the successive values of the (unpenalized) intercept. Equals to zero if intercept has been set to FALSE.

meanx:

Vector (class "numeric") containing the column means of the predictor matrix.

normx:

Vector (class "numeric") containing the square root of the sum of squares of each column of the design matrix.

penscale:

Vector "numeric" with real positive values that have been used to weight the penalty tuned by \lambda_1.

penalty:

Object of class "character" indicating the method used ("elastic-net" or "bounded regression").

naive:

logical; was the naive mode on?

lambda1:

Vector (class "numeric") of penalty levels (either \ell_1 or \ell_\infty) for which the model has eventually been fitted.

lambda2:

Scalar (class "numeric") for the amount of \ell_2 (ridge-like) penalty.

struct:

Object of class "Matrix" used to structure the coefficients in the \ell_2 penalty.

control:

Object of class "list" with low level options used for optimization.

monitoring:

List (class "list") which contains various indicators dealing with the optimization process.

residuals:

Matrix of residuals, each column corresponding to a value of lambda1.

r.squared:

Vector (class "numeric") given the coefficient of determination as a function of lambda1.

fitted:

Matrix of fitted values, each column corresponding to a value of lambda1.

Methods

This class comes with the usual predict(object, newx, ...), fitted(object, ...), residuals(object, ...), print(object, ...), show(object) and deviance(object, ...) generic (undocumented) methods.

A specific plotting method is available and documented (plot,quadrupen-method).

See Also

See also plot,quadrupen-method.