getDownSampleInd

This function selects the indices for the training set based on the class
vector <code>response</code>. It allows for optional downsampling to balance the class
distribution by limiting the number of samples per class.

A novel forward stepwise discriminant analysis framework
that integrates Pillai's trace with Uncorrelated Linear Discriminant Analysis (ULDA),
providing an improvement over traditional stepwise LDA methods that rely on Wilks' Lambda.
A stand-alone ULDA implementation is also provided, offering a more general solution
than the one available in the 'MASS' package. It automatically handles missing values and
provides visualization tools. For more details, see Wang (2024) <doi:10.48550/arXiv.2409.03136>.

Siyu Wang

folda

Forward Stepwise Discriminant Analysis with Pillai's Trace

getDownSampleInd function

<dl><dt>response</dt>
<dd>A factor vector representing the class labels.</dd>
<dt>downSampling</dt>
<dd>A logical value indicating whether downsampling should be
applied. If <code>TRUE</code>, downsampling is performed to limit the number of
samples per class based on <code>kSample</code>. Note that this may not result in
equal class frequencies, as <code>kSample</code> defines an upper limit for each
class, not a lower limit.</dd>
<dt>kSample</dt>
<dd>An integer specifying the maximum number of samples to be
selected per class. If <code>NULL</code>, the number of samples is limited to the size
of the smallest class.</dd></dl>

Arguments

Helper Function to Generate Training Set Indices Through Downsampling — getDownSampleInd

<dl>

<dt>response</dt>
<dd>A factor vector representing the class labels.</dd>


<dt>downSampling</dt>
<dd>A logical value indicating whether downsampling should be
applied. If <code>TRUE</code>, downsampling is performed to limit the number of
samples per class based on <code>kSample</code>. Note that this may not result in
equal class frequencies, as <code>kSample</code> defines an upper limit for each
class, not a lower limit.</dd>


<dt>kSample</dt>
<dd>An integer specifying the maximum number of samples to be
selected per class. If <code>NULL</code>, the number of samples is limited to the size
of the smallest class.</dd>

</dl>

getDownSampleInd: Helper Function to Generate Training Set Indices Through Downsampling

Description

Usage

Value

Arguments