simll

Parameter inference methods for models defined implicitly using a random simulator. Inference is carried out using simulation-based estimates of the log-likelihood of the data. The inference methods implemented in this package are explained in Park, J. (2025) <doi:10.48550/arxiv.2311.09446>. These methods are built on a simulation metamodel which assumes that the estimates of the log-likelihood are approximately normally distributed with the mean function that is locally quadratic around its maximum. Parameter estimation and uncertainty quantification can be carried out using the ht() function (for hypothesis testing) and the ci() function (for constructing a confidence interval for one-dimensional parameters).

Joonha Park

sbim

Simulation-Based Inference using a Metamodel for Log-Likelihood
Estimator

simll function

<dl><dt>ll</dt>
<dd>A matrix of simulation log likelihoods. The (i,m)-th entry is given by the simulation log likelihood for y_i obtained by simulating X at theta_m (e.g., the log density of y_i given X).</dd>
<dt>params</dt>
<dd>A matrix or a vector of parameter values. If a matrix, the m-th row gives the parameter vector theta_m. If theta is one dimensional, 'params' is can be a numeric vector or a matrix with one column. 'params' can be omitted if simulation log likelihoods are obtained at a single one parameter value.</dd>
<dt>weights</dt>
<dd>A numeric vector of weights, inversely proportional to the variance of simulation log likelihoods (optional)</dd></dl>

Arguments

Simulation Log Likelihood class — simll

<dl>

<dt>ll</dt>
<dd>A matrix of simulation log likelihoods. The (i,m)-th entry is given by the simulation log likelihood for y_i obtained by simulating X at theta_m (e.g., the log density of y_i given X).</dd>


<dt>params</dt>
<dd>A matrix or a vector of parameter values. If a matrix, the m-th row gives the parameter vector theta_m. If theta is one dimensional, 'params' is can be a numeric vector or a matrix with one column. 'params' can be omitted if simulation log likelihoods are obtained at a single one parameter value.</dd>


<dt>weights</dt>
<dd>A numeric vector of weights, inversely proportional to the variance of simulation log likelihoods (optional)</dd>

</dl>

simll: Simulation Log Likelihood class

Description

Usage

Value

Arguments