trans_boxcox

Applies the Box-Cox transformation to normalize the data distribution.
It automatically handles non-positive values by shifting the data.
The optimal lambda parameter is estimated using Maximum Likelihood Estimation (MLE).

Provides functions for data normalization and transformation in
preprocessing stages. Implements scaling methods (min-max, Z-score, L2
normalization) and power transformations (Box-Cox, Yeo-Johnson). Box-Cox
transformation is described in Box and Cox (1964)
<doi:10.1111/j.2517-6161.1964.tb00553.x>, Yeo-Johnson transformation in
Yeo and Johnson (2000) <doi:10.1093/biomet/87.4.954>.

Rui Gong

prepkit

Data Normalization and Transformation

trans_boxcox function

<dl><dt>x</dt>
<dd>A numeric vector.</dd>
<dt>lambda</dt>
<dd>A numeric value for the transformation power.
If <code>"auto"</code> (default), the optimal lambda is estimated within the interval [-2, 2].</dd>
<dt>force_pos</dt>
<dd>Logical. If <code>TRUE</code> (default), automatically shifts data to be positive
if non-positive values are present.</dd></dl>

Arguments

Box-Cox Transformation — trans_boxcox

<dl>

<dt>x</dt>
<dd>A numeric vector.</dd>


<dt>lambda</dt>
<dd>A numeric value for the transformation power.
If <code>"auto"</code> (default), the optimal lambda is estimated within the interval [-2, 2].</dd>


<dt>force_pos</dt>
<dd>Logical. If <code>TRUE</code> (default), automatically shifts data to be positive
if non-positive values are present.</dd>

</dl>

trans_boxcox: Box-Cox Transformation

Description

Usage

Value

Arguments

References