Portmanteau_test: Perform adjusted Portmanteau test for a STVAR model

In sstvars: Toolkit for Reduced Form and Structural Smooth Transition Vector Autoregressive Models

Perform adjusted Portmanteau test for a STVAR model

Description

Portmanteau_test performs adjusted Portmanteau test for remaining autocorrelation (or heteroskedasticity) in the residuals of a STVAR model.

Usage

Portmanteau_test(stvar, nlags = 20, which_test = c("autocorr", "het.sked"))

Arguments

stvar	an object of class 'stvar' generated by fitSTVAR or STVAR.
nlags	a strictly positive integer specifying the number of lags to be tested.
which_test	should test for remaining autocorrelation or heteroskedasticity be calculated?

Details

The implemented adjusted Portmanteau test is based on Lütkepohl (2005), Section 4.4.3. When testing for remaining heteroskedasticity, the Portmanteau test is applied to squared standardized residuals that are centered to have zero mean. Note that the validity of the heteroskedasticity test requires that the residuals are not autocorrelated.

Value

A list with class "hypotest" containing the test results and arguments used to calculate the test.

References

• Lütkepohl H. 2005. New Introduction to Multiple Time Series Analysis, Springer.

See Also

LR_test, Rao_test, fitSTVAR, STVAR, diagnostic_plot, profile_logliks,

Examples

# Gaussian STVAR p=2, M=2, model with weighted relative stationary densities
# of the regimes as the transition weight function:
theta_222relg <- c(0.357, 0.107, 0.356, 0.086, 0.14, 0.035, -0.165, 0.387, 0.452,
 0.013, 0.228, 0.336, 0.239, 0.024, -0.021, 0.708, 0.063, 0.027, 0.009, 0.197,
 0.206, 0.005, 0.026, 1.092, -0.009, 0.116, 0.592)
mod222relg <- STVAR(data=gdpdef, p=2, M=2, d=2, params=theta_222relg,
 weight_function="relative_dens")

# Test for remaining autocorrelation taking into account the first 20 lags:
Portmanteau_test(mod222relg, nlags=20)

# Test for remaining heteroskedasticity taking into account the first 20 lags:
Portmanteau_test(mod222relg, nlags=20, which_test="het.sked")

# Two-regime Student's t Threhold VAR p=3 model with the first lag of the second
# variable as the switching variable:
theta_322thres <- c(0.527, 0.039, 1.922, 0.154, 0.284, 0.053, 0.033, 0.453, 0.291,
 0.024, -0.108, 0.153, -0.108, 0.003, -0.128, 0.219, 0.195, -0.03, -0.893, 0.686,
 0.047, 0.016, 0.524, 0.068, -0.025, 0.044, -0.435, 0.119, 0.359, 0.002, 0.038,
 1.252, -0.041, 0.151, 1.196, 12.312)
mod322thres <- STVAR(data=gdpdef, p=3, M=2, d=2, params=theta_322thres,
 weight_function="threshold", weightfun_pars=c(2, 1), cond_dist="Student")

# Test for remaining autocorrelation taking into account the first 25 lags:
Portmanteau_test(mod322thres, nlags=25)

# Test for remaining heteroskedasticity taking into account the first 25 lags:
Portmanteau_test(mod322thres, nlags=25, which_test="het.sked")

sstvars documentation built on April 11, 2025, 5:47 p.m.