eval_on_grid_sim_model: Evaluate Expected Utility On A Grid
In bdpopt: Optimisation of Bayesian Decision Problems

eval.on.grid method for objects of class sim.model.

1
2
3

## S3 method for class 'sim.model'
eval.on.grid(model, utility.fun, grid.spec.list,
n.iter, n.burn.in, n.adapt = 1000, independent.SE = FALSE, parallel = FALSE)

`model`	Simulation model object of class `sim.model` created using `sim.model` or `create.normal.model`.
`utility.fun`	An R function defining the utility for the decision problem. It must be possible to extract the argument names of the function supplied with `formals`, and the argument names must constitute a subset of the names used in the JAGS model specification.
`grid.spec.list`	A nonempty list of array grid specifications. An array grid specification is a list of two componenents. The first component is a dimension vector, giving the dimensions of the array. The second component is a list of vectors of length equal to the product of the dimension vector. Each such vector has the form `c(lower, upper, step)`. These are passed to the function `seq` in order to generate a range of values for each component of the array.
`n.iter`	The number of iterations in the JAGS MCMC simulation for each grid point.
`n.burn.in`	The number of burn in iterations prior to the JAGS MCMC simulation for each grid point.
`n.adapt`	The number of adaptation iterations prior to the burn in phase and subsequent JAGS MCMC simulation for each grid point.
`independent.SE`	If `TRUE`, then the standard errors of the sample means used to estimate the expected utility will be computed under the assumption of i.i.d. sampling. If `FALSE`, the standard errors are instead computed using the `coda::spectrum0.ar` function.
`parallel`	Logical equal to `TRUE` if the simulation should be done in parallel on a multi-core processor. The default value `FALSE` leads to single-core evaluation.