This function computes the data-driven bandwidth for smoothing the ROC (or distribution) function using the PI method of Beyene and El Ghouch (2020). This is an extension of the classical (unweighted) direct plug-in bandwith selection method to the case of weighted data.

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

PI function

<dl><dt>X</dt>
<dd>The numeric vector of random variable.</dd>
<dt>wt</dt>
<dd>The non-negative weight vector.</dd>
<dt>ktype</dt>
<dd>A character string giving the type kernel to be used: "<code>normal</code>", "<code>epanechnikov</code>", "<code>biweight</code>", or "<code>triweight</code>". By default, the "<code>normal</code>" kernel is used.</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

The plug-in bandwidth selection for weighted data — PI

<dl>

<dt>X</dt>
<dd>The numeric vector of random variable.</dd>


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


<dt>ktype</dt>
<dd>A character string giving the type kernel to be used: "<code>normal</code>", "<code>epanechnikov</code>", "<code>biweight</code>", or "<code>triweight</code>". By default, the "<code>normal</code>" kernel is used.</dd>

</dl>

PI: The plug-in bandwidth selection for weighted data

Description

Usage

Value

Arguments

Author

Details

References