star: STAR model

Description Usage Arguments Details Value Author(s) See Also Examples

View source: R/star.R

Description

STAR model fitting with automatic selection of the number of regimes based on LM tests.

Usage

1
2
star(x, m=2, noRegimes, d = 1, steps = d, series, rob = FALSE,
                 mTh, thDelay, thVar, sig=0.05, trace=TRUE, control=list(), ...)

Arguments

x

time series

m, d, steps

embedding dimension, time delay, forecasting steps

noRegimes

max number of regimes

series

time series name (optional)

rob

perform robust test (not implemented)

thDelay

'time delay' for the threshold variable (as multiple of embedding time delay d)

mTh

coefficients for the lagged time series, to obtain the threshold variable

thVar

external threshold variable

sig

significance level for the tests to select the number of regimes.

control

further arguments to be passed as control list to optim

trace

should additional infos be printed out?

...

currently unused

Details

The function star implements the iterative building strategy described in [1] to identify and estimate Smooth Transition AutoRegressive models.

[1] T. Terasvirta, "Specification, estimation and evaluation of smooth transition autoregresive models", J. Am. Stat. Assoc. 89 (1994): 208-218.

Value

star returns an object of class nlar, subclass star, i.e. a list with informations about the fitted model.

Author(s)

J. L. Aznarte M.

See Also

addRegime

Examples

1
2
3
4
mod.star <- star(log10(lynx), mTh=c(0,1), control=list(maxit=3000))
mod.star

addRegime(mod.star)

Example output

Testing linearity...   p-Value =  0.0001831653 
The series is nonlinear. Incremental building procedure:
Building a 2 regime STAR.
Performing grid search for starting values...
Starting values fixed: gamma =  11.15385 , th =  3.337486 ; SSE =  4.337664 
Optimization algorithm converged
Optimized values fixed for regime 2  : gamma =  11.15383 , th =  3.339199 ; SSE =  4.337643 

Testing for addition of regime 3.
  Estimating gradient matrix...
  Done. Computing the test statistic...
  Done. Regime 3 is NOT accepted (p-Value =  0.9843362 ).

Finished building a MRSTAR with  2  regimes

Non linear autoregressive model

Multiple regime STAR model

Regime  1 :
    Linear parameters: 0.4891014, 1.2465399, -0.3664328 

Regime  2 :
    Linear parameters: -1.0240758, 0.4232669, -0.2546088 
    Non-linear parameters:
11.1538344, 3.3391985

Non linear autoregressive model

Multiple regime STAR model

Regime  1 :
    Linear parameters: 0.4891014, 1.2465399, -0.3664328 

Regime  2 :
    Linear parameters: -1.0240758, 0.4232669, -0.2546088 
    Non-linear parameters:
11.1538344, 3.3391985
Regime  3 :
    Linear parameters: -0.5808191, -0.1599709, -2.9181273 
    Non-linear parameters:
0, 0

tsDyn documentation built on May 29, 2017, 10:48 a.m.