makeTSnoise: Generate a random time series

Description Usage Arguments Value

Description

Generate a random time series (matrix). This is a utility typically used in a time series model simulate method and not called directly by the user.

Usage

1
2
3
    makeTSnoise(sampleT,p,lags,noise=NULL, rng=NULL, Cov=NULL, sd=1, 
	     noise.model=NULL, noise.baseline=0,
             tf=NULL, start=NULL,frequency=NULL)

Arguments

sampleT

an integer indicating the number of periods.

p

an integer indicating the number of series.

lags

an integer indicating the number of periods prior to the sample (initial data w0) for which random numbers should be generated. This is useful in ARMA models.

noise

Noise can be supplied. Otherwise it will be generated. If supplied it should be a list as described below under returned value.

Cov

The covariance of the noise process. If this is specified then sd is ignored. A vector or scalar is treated as a diagonal matrix. For an object of class TSestModel, if neither Cov nor sd are specified, then Cov is set to the estimated covariance (model$estimates$cov).

sd

The standard deviation of the noise. This can be a vector.

noise.model

A TSmodel to be used for generating noise (not yet supported by SS methods).

noise.baseline

a constant or matrix to be added to noise. Alternately this can be a vector of length p, each value of which is treated as a constant to add to the coresponding noise series.

rng

The random number generator information needed to regenerate a simulation.

tf

a time frame to use for the generated matrix. (alternately use start and frequency)

start

a time start date to use for the generated matrix.

frequency

a time frequency to use for the generated matrix.

Value

A time series matrix.


dse documentation built on March 26, 2020, 7:12 p.m.