projection_depth

Helper function to compute the random projection depth of multivariate point(s) with respect
to a multivariate data.

A collection of functions for outlier detection in functional data analysis.
Methods implemented include directional outlyingness by
Dai and Genton (2019) <doi:10.1016/j.csda.2018.03.017>,
MS-plot by Dai and Genton (2018) <doi:10.1080/10618600.2018.1473781>,
total variation depth and modified shape similarity index by
Huang and Sun (2019) <doi:10.1080/00401706.2019.1574241>, and sequential transformations by
Dai et al. (2020) <doi:10.1016/j.csda.2020.106960 among others. Additional outlier detection
tools and depths for functional data like functional boxplot, (modified) band depth etc.,
are also available.

Oluwasegun Taiwo Ojo

fdaoutlier

Outlier Detection Tools for Functional Data Analysis

Rosa Elvira Lillo

Antonio Fernandez Anta

projection_depth function

<dl><dt>dts</dt>
<dd>A matrix or data frame of size <code>m</code> observations by <code>d</code> dimension or vector of length <code>d</code>.
Contains the observation(s) whose depth is to be computed.</dd>
<dt>dt</dt>
<dd>A matrix or dataframe of size <code>n</code> observations by <code>d</code> dimension. Equals to
<code>dts</code> by default.</dd>
<dt>n_projections</dt>
<dd>The number of directions for random projections. By default,
500 random directions for projection are generated from a scaled uniform distribution between -1 and 1.</dd>
<dt>seed</dt>
<dd>The random seed to set when generating the random directions. Defaults to NULL.</dd></dl>

Arguments

Author

Random projection for multivariate data — projection_depth

<dl>

<dt>dts</dt>
<dd>A matrix or data frame of size <code>m</code> observations by <code>d</code> dimension or vector of length <code>d</code>.
Contains the observation(s) whose depth is to be computed.</dd>


<dt>dt</dt>
<dd>A matrix or dataframe of size <code>n</code> observations by <code>d</code> dimension. Equals to
<code>dts</code> by default.</dd>


<dt>n_projections</dt>
<dd>The number of directions for random projections. By default,
500 random directions for projection are generated from a scaled uniform distribution between -1 and 1.</dd>


<dt>seed</dt>
<dd>The random seed to set when generating the random directions. Defaults to NULL.</dd>

</dl>

projection_depth: Random projection for multivariate data

Description

Usage

Value

Arguments

Author

See Also

Examples