starma_stat_check: Check a STARMA model for stationarity

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

Description

Checks the coefficients of a STARMA models for stationarity.

Usage

starma_stat_check(model)

Arguments

model

A list with components ar and ma. Each component contains a matrix of the coefficients as specified in this functions' details.

Details

The stationarity constraints for a STAR model are the following:

• STAR(2_11)

  - phi_20 + abs(phi_21) <1

  abs(phi_10 + phi_11) < 1 - phi_20 - phi_21

  abs(phi_10 - phi_11) < 1 - phi_20 + phi_21

• STAR(2_10), meaning phi_21 = 0

  phi_20 > -1

  abs(phi_10) + abs(phi_11) < 1 - phi_20

• STAR(2_00), meaning phi_21 = 0 \& phi_11 = 0

  phi_20 > -1

  abs(phi_10) < 1 - phi_20

• STAR(1_1), meaning phi_20 = 0 \& phi_21 = 0

  abs(phi_10) + abs(phi_11) < 1

• STAR(1_0), meaning phi=20 = 0 \& phi_21 = 0 \& phi_11 = 0

  abs(phi_10) < 1

where phi_10 = model$ar[1,1], phi_11 = model$ar[1,2], phi_20 = model$ar[2,1], and phi_21 = model$ar[2,2]

STMA models have the same constraints for theta as the STAR contraints for phi.

Value

TRUE if the model is stationary. FALSE, otherwise.

References

http://www.tandfonline.com/doi/abs/10.1080/03610918008812173

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