simul

Performs Gaussian process regression with heteroskedastic noise following the model by Binois, M., Gramacy, R., Ludkovski, M. (2016) <doi:10.48550/arXiv.1611.05902>, with implementation details in Binois, M. & Gramacy, R. B. (2021) <doi:10.18637/jss.v098.i13>. The input dependent noise is modeled as another Gaussian process. Replicated observations are encouraged as they yield computational savings. Sequential design procedures based on the integrated mean square prediction error and lookahead heuristics are provided, and notably fast update functions when adding new observations.

Mickael Binois

hetGP

Heteroskedastic Gaussian Process Modeling and Design under
Replication

simul function

<dl><dt>object</dt>
<dd><code>CRNGP</code> model</dd>
<dt>Xgrid</dt>
<dd>matrix of (x, seed) locations where the simulation is performed. 
Where all design locations are matched with all seed values. In particular, it is assumed that each unique x values is matched with all seeds before going to the next x value.
The last column MUST correspond to seeds values. <code>Xgrid</code> must also contain the evaluated designs (e.g., in model$X0)</dd>
<dt>ids</dt>
<dd>vector of indices corresponding to observed values in <code>Xgrid</code></dd>
<dt>nsim</dt>
<dd>number of simulations to return</dd>
<dt>eps</dt>
<dd>jitter used in the Cholesky decomposition of the covariance matrix for numerical stability</dd>
<dt>seqseeds</dt>
<dd>is the seed sequence repeated (e.g., 1 2 3 1 2 3), else it is assumed to be ordered (e.g., 1 1 2 2 3 3)</dd>
<dt>check</dt>
<dd>if <code>TRUE</code>, check that Xgrid has the proper structure (slower)</dd></dl>

Arguments

Conditional simulation for CRNGP — simul

<dl>

<dt>object</dt>
<dd><code>CRNGP</code> model</dd>


<dt>Xgrid</dt>
<dd>matrix of (x, seed) locations where the simulation is performed. 
Where all design locations are matched with all seed values. In particular, it is assumed that each unique x values is matched with all seeds before going to the next x value.
The last column MUST correspond to seeds values. <code>Xgrid</code> must also contain the evaluated designs (e.g., in model$X0)</dd>


<dt>ids</dt>
<dd>vector of indices corresponding to observed values in <code>Xgrid</code></dd>


<dt>nsim</dt>
<dd>number of simulations to return</dd>


<dt>eps</dt>
<dd>jitter used in the Cholesky decomposition of the covariance matrix for numerical stability</dd>


<dt>seqseeds</dt>
<dd>is the seed sequence repeated (e.g., 1 2 3 1 2 3), else it is assumed to be ordered (e.g., 1 1 2 2 3 3)</dd>


<dt>check</dt>
<dd>if <code>TRUE</code>, check that Xgrid has the proper structure (slower)</dd>

</dl>

simul: Conditional simulation for CRNGP

Description

Usage

Value

Arguments