weighted.quantile

<code>weighted.quantile()</code> produces weighted sample quantiles corresponding to the given probabilities.

internal

The goal of 'midr' is to provide a model-agnostic method for interpreting and explaining black-box predictive models by creating a globally interpretable surrogate model. The package implements 'Maximum Interpretation Decomposition' (MID), a functional decomposition technique that finds an optimal additive approximation of the original model. This approximation is achieved by minimizing the squared error between the predictions of the black-box model and the surrogate model. The theoretical foundations of MID are described in Iwasawa & Matsumori (2025) [Forthcoming], and the package itself is detailed in Asashiba et al. (2025) <doi:10.48550/arXiv.2506.08338>.

Ryoichi Asasihba

midr

Learning from Black-Box Models by Maximum Interpretation
Decomposition

Ryoichi Asashiba

Hirokazu Iwasawa

Reiji Kozuma

weighted.quantile function

<dl><dt>x</dt>
<dd>a numeric vector whose weighted sample quantiles are wanted.</dd>
<dt>w</dt>
<dd>a numeric vector of the sample weights for each value in <code>x</code>.</dd>
<dt>probs</dt>
<dd>a numeric vector of probabilities with values in <code>[0, 1]</code>.</dd>
<dt>na.rm</dt>
<dd>logical. If <code>TRUE</code>, any <code>NA</code> and <code>NaN</code>s are removed from <code>x</code> before the quantiles are computed.</dd>
<dt>names</dt>
<dd>logical. If <code>TRUE</code>, the result has a "names" attribute.</dd>
<dt>digits</dt>
<dd>used only when <code>names</code> is <code>TRUE</code>. The precision to use when formatting the percentages.</dd>
<dt>type</dt>
<dd>an integer between <code>1</code> and <code>9</code> selecting the quantile algorithms. Only <code>1</code> is available for the weighted quantile.</dd>
<dt>...</dt>
<dd>further arguments passed to <code><a href="/link/stats%3A%3Aquantile()?package=midr&version=0.5.2" data-mini-rdoc="midr::stats::quantile()">stats::quantile()</a></code> when weights are not passed.</dd></dl>

Arguments

Weighted Sample Quantile — weighted.quantile

<dl>

<dt>x</dt>
<dd>a numeric vector whose weighted sample quantiles are wanted.</dd>


<dt>w</dt>
<dd>a numeric vector of the sample weights for each value in <code>x</code>.</dd>


<dt>probs</dt>
<dd>a numeric vector of probabilities with values in <code>[0, 1]</code>.</dd>


<dt>na.rm</dt>
<dd>logical. If <code>TRUE</code>, any <code>NA</code> and <code>NaN</code>s are removed from <code>x</code> before the quantiles are computed.</dd>


<dt>names</dt>
<dd>logical. If <code>TRUE</code>, the result has a "names" attribute.</dd>


<dt>digits</dt>
<dd>used only when <code>names</code> is <code>TRUE</code>. The precision to use when formatting the percentages.</dd>


<dt>type</dt>
<dd>an integer between <code>1</code> and <code>9</code> selecting the quantile algorithms. Only <code>1</code> is available for the weighted quantile.</dd>


<dt>...</dt>
<dd>further arguments passed to <code><a href='https://rdrr.io/r/stats/quantile.html'>stats::quantile()</a></code> when weights are not passed.</dd>

</dl>

weighted.quantile: Weighted Sample Quantile

Description

Usage

Value

Arguments

Details

See Also