stsm-class: Class 'stsm' for Structural Time Series Models

In stsm: Structural Time Series Models

Description

This class defines a structural time series model.

Slots

call

Object of class call. Call to stsm.model.

model

Object of class character. Name or label for the selected model (see stsm.model for available models).

y

Object of class ts. Original time series.

diffy

Object of class ts. Differenced series y. The differencing operator that renders stationarity in the model is applied to the series y.

xreg

An optinal matrix or numeric vector of external regressors.

fdiff

Object of class function. Function with arguments x: a ts object, s: periodicity of the data. This function applies the differencing operator that renders stationarity in the model to a ts object passed to it.

ss

Object of class list. Matrices of the state space form of the structural model.

pars

Object of class numeric. Named vector with the parameters of the model.

nopars

Optional object of class numeric. An optional named vector with the remaining parameters of the model not included in pars. This slot is not affected by the transformation of parameters transPars. These parameters are considered fixed in the optimization procedures implemented in package stsm.

cpar

Optional object of class numeric. Named vector of length one containing the parameter that is concentrated out of the likelihood function (if any).

lower

Object of class numeric. Named vector with the lower bounds for pars.

upper

Object of class numeric. Named vector with the upper bounds for pars.

transPars

Character string referring to the parameterization of the model, see transPars.

ssd

Optional object of class numeric. Sample spectral density (periodogram) of the differenced series diffy.

sgfc

Optional object of class matrix. Constant elements in the spectral generating function of the model (for pure variance models).

Methods

char2numeric

Return a list containing the matrices of the state space representation of the model. The matrices are the same as those in the slot ss but the characters are replaced by the corresponding numeric values defined in pars, nopars and cpar.

checkbounds

Check whether the values of pars lie within the lower and upper bounds.

get.pars

Return the slot pars, the parameters of the model. If the model is parameterized in terms of a set of auxiliary parameters such as those considered in transPars, then the transformed parameters are returned. Thus, when the slot transPars is not NULL x@pars will not be equal to get.pars(x).

get.cpar

Return the slot cpar.

get.nopars

Return the slot nopars.

set.cpar

Set or modify the value of the slot cpar

set.nopars

Set or modify the value of the slot nopars.

set.pars

Set or modify the value of the slot pars.

set.sgfc

Compute and set the value of the slot sgfc.

set.xreg

Set or modify the value of the slot xreg.

setValidity

Check the validity of the arguments passed to the function.

show

Show a brief summary of the object.

transPars

Transform the parameters of the model according to the parameterization defined in the slot transPars.

References

Christophe Genolini. A (Not So) Short Introduction to S4. Object Oriented Programming in R. V0.5.1. August 20, 2008.

See Also

stsm.model.

stsm documentation built on May 2, 2019, 7:39 a.m.