sobol4r_run

Helper around <code>sensitivity::sobol</code> that mimics the structure
of the original scripts. It never writes to disk.

Tools to design experiments, compute Sobol sensitivity indices,
and summarise stochastic responses inspired by the strategy described by
Zhu and Sudret (2021) <doi:10.1016/j.ress.2021.107815>. Includes helpers
to optimise toy models implemented in C++, visualise indices with
uncertainty quantification, and derive reliability-oriented sensitivity
measures based on failure probabilities.
It is further detailed in Logosha, Maumy and Bertrand (2022)
<doi:10.1063/5.0246026> and (2023) <doi:10.1063/5.0246024> or in Bertrand,
Logosha and Maumy (2024) <https://hal.science/hal-05371803>,
<https://hal.science/hal-05371795> and <https://hal.science/hal-05371798>.

Frederic Bertrand

Sobol4R

Sobol Indices for Models with Fixed and Stochastic Parameters

Elizaveta Logosha

Myriam Maumy-Bertrand

sobol4r_run function

<dl><dt>model</dt>
<dd>Deterministic or stochastic model that takes a design X
and returns a numeric vector of length nrow(X).</dd>
<dt>X1, X2</dt>
<dd>Matrices or data.frames used to build the Sobol design.</dd>
<dt>order</dt>
<dd>Order of the Sobol indices (1 or 2).</dd>
<dt>nboot</dt>
<dd>Number of bootstrap replicates for confidence intervals.</dd>
<dt>qoi_fun</dt>
<dd>Optional quantity of interest function. If not NULL,
the model is evaluated repeatedly and QoI is computed row wise.</dd>
<dt>nrep</dt>
<dd>Number of replications per design row when <code>qoi</code> is not NULL.</dd>
<dt>type</dt>
<dd>Type of Monte Carlo Estimation of Sobol' Indices to be used.
Supported estimators mirror the sensitivity helpers: <code>sobol</code>,
<code>sobol2007</code>, <code>soboljansen</code>, <code>sobolEff</code>, and
<code>sobolmartinez</code>. Defaults to <code>"soboljansen"</code> because it offers
robust first and total order indices on both centred and non-centred
outputs.</dd>
<dt>...</dt>
<dd>Extra arguments passed to <code>model</code>.</dd></dl>

Arguments

Run Sobol analysis with optional QoI wrapper — sobol4r_run

<dl>

<dt>model</dt>
<dd>Deterministic or stochastic model that takes a design X
and returns a numeric vector of length nrow(X).</dd>


<dt>X1, X2</dt>
<dd>Matrices or data.frames used to build the Sobol design.</dd>


<dt>order</dt>
<dd>Order of the Sobol indices (1 or 2).</dd>


<dt>nboot</dt>
<dd>Number of bootstrap replicates for confidence intervals.</dd>


<dt>qoi_fun</dt>
<dd>Optional quantity of interest function. If not NULL,
the model is evaluated repeatedly and QoI is computed row wise.</dd>


<dt>nrep</dt>
<dd>Number of replications per design row when <code>qoi</code> is not NULL.</dd>


<dt>type</dt>
<dd>Type of Monte Carlo Estimation of Sobol' Indices to be used.
Supported estimators mirror the sensitivity helpers: <code>sobol</code>,
<code>sobol2007</code>, <code>soboljansen</code>, <code>sobolEff</code>, and
<code>sobolmartinez</code>. Defaults to <code>"soboljansen"</code> because it offers
robust first and total order indices on both centred and non-centred
outputs.</dd>


<dt>...</dt>
<dd>Extra arguments passed to <code>model</code>.</dd>

</dl>

sobol4r_run: Run Sobol analysis with optional QoI wrapper

Description

Usage

Value

Arguments