monte.carlo.simulations {dse2} | R Documentation |
Run multiple simulations
is.monte.carlo.simulation(obj) monte.carlo.simulations(model, simulation.args=NULL, replications=100, rng=NULL, ...) monte.carlo.simulations(model, simulation.args = NULL, replications = 100, rng = NULL, quiet = FALSE) monte.carlo.simulations(model, simulation.args=NULL, replications=100, rng=NULL, Spawn=.SPAWN, quiet=F) monte.carlo.simulations(model, simulation.args=NULL, replications=100, rng=NULL, ...) monte.carlo.simulations(model,...) monte.carlo.simulations(model,...)
model |
A model with a simulate 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. |
Spawn |
If T "For" loops are used in Splus. |
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 estimation.evaluation
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.
A list of simulations.
simulate
eval.estimation
distribution
forecast.cov.wrt.true
if(is.R()) data("eg1.DSE.data.diff", package="dse1") model <- est.VARX.ls(eg1.DSE.data.diff) z <- monte.carlo.simulations(model, simulation.args=list(sampleT=100)) tfplot(z) distribution(z)