starma_sim: Simulate spatio-temporal data using a STARMA model

In mrfoliveira/Evaluation-procedures-for-forecasting-with-spatio-temporal-data: Evaluation procedures for forecasting with spatio-temporal data

Description

Generate a spatio-teporal dataset according to a STARMA model.

Usage

starma_sim(model, klist, n, rand.gen = stats::rnorm, innov = NULL,
  seed = NULL, FUN = function(x) {     x }, ...)

Arguments

model	A list with components ar and ma. Each component contains a matrix of the coefficients where the first index corresponds to the row and the second index to the column.
klist	A list of matrices like the ones returned by consecutive use of spdep::dnearneigh and spdep::nblag where a value higher than 0 implies that the row and column locations are neighbours
n	The length of the time series to be generated
rand.gen	The random generator to be used. Defaults to stats::rnorm
innov	A matrix of initial innovations. Defaults to matrix(rand.gen(n*ncol(klist[[1]]), ...), n, ncol(klist[[1]]))
seed	A seed to set before generating the dataset. Defaults to NULL
FUN	A (possibly non-linear) function to apply to the matrix during STAR data generation.
...	Other parameters (?)

Value

A matrix where each column contains the data for a location, and each row contains the data for a time-stamp.

References

See example on page 3 of https://cran.r-project.org/web/packages/starma/starma.pdf

mrfoliveira/Evaluation-procedures-for-forecasting-with-spatio-temporal-data documentation built on April 11, 2021, 10:50 a.m.