simData

The package includes a simulated dataset that we will use for the examples.

datasets

Fit a model with potentially many linear and smooth predictors. Interaction
effects can also be quantified. Variable selection is done using penalisation. For l1-type penalties
we use iterative steps alternating between using linear predictors (lasso) and smooth predictors
(generalised additive model).

Indrayudh Ghosal

plsmselect

Linear and Smooth Predictor Modelling with Penalisation and
Variable Selection

Matthias Kormaksson

simData function

A 100-by-23 data frame. There are 10 variables
 (<code>x1</code>,...,<code>x10</code>) corresponding to the linear predictors and 4
 (<code>z1</code>,...,<code>z4</code>) corresponding to the smooth predictors. There are
 7 response variables corresponding to the different models fitted -<ul>
<li><code>Yg</code> for the Gaussian response</li>
<li><code>Yb</code> as Bernoulli and <code>success</code> and <code>failure</code> as
 Binomial count responses</li>
<li><code>Yp</code> as the Poisson response</li>
<li><code>time</code> and <code>status</code> as the survival model responses</li>
</ul>The variables starting with <code>X</code> are the same as the linear predictors but are
 concatenated into a matrix <code>X</code> to be used for the formula implementation of
 <code>gamlasso</code>

simData: Simulated dataset to be used for gamlasso

Description

Usage

Arguments

Format

Details

Examples