This function computes the data-driven bandwidth value for smoothing the ROC curve.
It contains three methods: the normal refrence, the plug-in and the cross-validation methods.

internal

This software package provides Cox survival analysis for high-dimensional and multiblock datasets.
It encompasses a suite of functions dedicated from the classical Cox regression to newest analysis,
including Cox proportional hazards model, Stepwise Cox regression, and Elastic-Net Cox regression,
Sparse Partial Least Squares Cox regression (sPLS-COX) incorporating three distinct strategies,
and two Multiblock-PLS Cox regression (MB-sPLS-COX) methods. This tool is designed to adeptly handle
high-dimensional data, and provides tools for cross-validation, plot generation, and additional resources
for interpreting results. While references are available within the corresponding functions,
key literature is mentioned below.
Terry M Therneau (2024) <https://CRAN.R-project.org/package=survival>,
Noah Simon et al. (2011) <doi:10.18637/jss.v039.i05>,
Philippe Bastien et al. (2005) <doi:10.1016/j.csda.2004.02.005>,
Philippe Bastien (2008) <doi:10.1016/j.chemolab.2007.09.009>,
Philippe Bastien et al. (2014) <doi:10.1093/bioinformatics/btu660>,
Kassu Mehari Beyene and Anouar El Ghouch (2020) <doi:10.1002/sim.8671>,
Florian Rohart et al. (2017) <doi:10.1371/journal.pcbi.1005752>.

Pedro Salguero Garcia

Coxmos

Cox MultiBlock Survival

Pedro Salguero García

Sonia Tarazona Campos

Ana Conesa Cegarra

Kassu Mehari Beyene

Luis Meira Machado

Marta Sestelo

Artur Araújo

wbw function

<dl><dt>X</dt>
<dd>The numeric data vector.</dd>
<dt>wt</dt>
<dd>The non-negative weight vector.</dd>
<dt>bw</dt>
<dd>A character string specifying the bandwidth selection method. The possible options are "<code>NR</code>" for the normal reference, the plug-in "<code>PI</code>" and cross-validation "<code>CV</code>".</dd>
<dt>ktype</dt>
<dd>A character string indicating the type of kernel function: "<code>normal</code>", "<code>epanechnikov</code>", "<code>biweight</code>", or "<code>triweight</code>". Default is "<code>normal</code>" kernel.</dd></dl>

Arguments

Kassu Mehari Beyene, Catholic University of Louvain. <code>&lt;kasu.beyene@uclouvain.be&gt;</code>
Anouar El Ghouch, Catholic University of Louvain. <code>&lt;anouar.elghouch@uclouvain.be&gt;</code>

Author

Function to select the bandwidth parameter needed for smoothing the time-dependent ROC curve. — wbw

<dl>

<dt>X</dt>
<dd>The numeric data vector.</dd>


<dt>wt</dt>
<dd>The non-negative weight vector.</dd>


<dt>bw</dt>
<dd>A character string specifying the bandwidth selection method. The possible options are "<code>NR</code>" for the normal reference, the plug-in "<code>PI</code>" and cross-validation "<code>CV</code>".</dd>


<dt>ktype</dt>
<dd>A character string indicating the type of kernel function: "<code>normal</code>", "<code>epanechnikov</code>", "<code>biweight</code>", or "<code>triweight</code>". Default is "<code>normal</code>" kernel.</dd>

</dl>

wbw: Function to select the bandwidth parameter needed for smoothing the time-dependent ROC curve.

Description

Usage

Value

Arguments

Author

References