dse (version 2007.7-1)

MonteCarloSimulations: Generate simulations

Description

Run multiple simulations

Usage

is.MonteCarloSimulations(obj)
    MonteCarloSimulations(model, simulation.args=NULL, 
           replications=100, rng=NULL, quiet =FALSE, ...)
    ## S3 method for class 'default':
MonteCarloSimulations(model, simulation.args = NULL, 
 		replications = 100, rng = NULL, quiet =FALSE, ...)
    ## S3 method for class 'TSmodel':
MonteCarloSimulations(model, simulation.args=NULL,
          replications=100, rng=NULL, quiet=FALSE, ...)    ## S3 method for class 'TSestModel':
MonteCarloSimulations(model, simulation.args=NULL, 
           replications=100, rng=NULL, quiet=FALSE, ...)
    ## S3 method for class 'EstEval':
MonteCarloSimulations(model, simulation.args=NULL,
            replications=100, rng=getRNG(model),  quiet=FALSE, ...)
    ## S3 method for class 'MonteCarloSimulations':
MonteCarloSimulations(model, 
       simulation.args=NULL, replications=100, rng=getRNG(model),  quiet=FALSE, ...)

Arguments

model

an object from which a model can be extracted. The model must have an associated simulation method (e.g. a TSmodel).

simulation.args,

A list of arguments in addition to model which are passed to simulate.

replications

The number of simulations.

rng

The RNG and starting seed.

quiet

logical indicating if printing and many warning messages should be suppressed.

obj

an object.

...

arguments passed to other methods.

Value

• A list of simulations.

concept

DSE

Details

This function runs many simulations using simulate. Often it not be necessary to do this since the seed can be used to reproduce the sample and many functions for testing estimation methods, etc., will produce samples as they proceed. This function is useful for verification and for looking at the stochastic properties of the output of a model. If model is an object of class EstEval or simulation then the model and the seed!!! are extracted so the same sample will be generated. The default method expects the result of simulate(model) to be a matrix. There is a tfplot method (time series plots of the simulations) and a distribution method for the result. The latter plots kernel estimates of the distribution of the simulations at specified periods.

See Also

simulate EstEval distribution forecastCovWRTtrue

Examples

data("eg1.DSE.data.diff", package="dse1")
model <- estVARXls(eg1.DSE.data.diff)
z <-  MonteCarloSimulations(model, simulation.args=list(sampleT=100))
tfplot(z)
distribution(z)