Conduct Wald Test
This example shows how to calculate the required inputs for conducting a Wald test with waldtest
. The Wald test compares the fit of a restricted model against an unrestricted model by testing whether the restriction function, evaluated at the unrestricted maximum likelihood estimates (MLEs), is significantly different from zero.
The required inputs for waldtest
are a restriction function, the Jacobian of the restriction function evaluated at the unrestricted MLEs, and an estimate of the variance-covariance matrix evaluated at the unrestricted MLEs. This example compares the fit of an AR(1) model against an AR(2) model.
Compute Unrestricted MLE
Obtain the unrestricted MLEs by fitting an AR(2) model (with a Gaussian innovation distribution) to the given data. Assume you have presample observations () = (9.6249,9.6396)
Y = [10.1591; 10.1675; 10.1957; 10.6558; 10.2243; 10.4429;
10.5965; 10.3848; 10.3972; 9.9478; 9.6402; 9.7761;
10.0357; 10.8202; 10.3668; 10.3980; 10.2892; 9.6310;
9.6318; 9.1378; 9.6318; 9.1378];
Y0 = [9.6249; 9.6396];
Mdl = arima(2,0,0);
[EstMdl,V] = estimate(Mdl,Y,'Y0',Y0);
ARIMA(2,0,0) Model (Gaussian Distribution): Value StandardError TStatistic PValue _______ _____________ __________ _________ Constant 2.8802 2.5239 1.1412 0.25379 AR{1} 0.60623 0.40372 1.5016 0.1332 AR{2} 0.10631 0.29283 0.36303 0.71658 Variance 0.12386 0.042598 2.9076 0.0036425
When conducting a Wald test, only the unrestricted model needs to be fit. estimate
returns the estimated variance-covariance matrix as an optional output.
Compute Jacobian Matrix
Define the restriction function, and calculate its Jacobian matrix.
For comparing an AR(1) model to an AR(2) model, the restriction function is
The Jacobian of the restriction function is
Evaluate the restriction function and Jacobian at the unrestricted MLEs.
r = EstMdl.AR{2}; R = [0 0 1 0];
Conduct Wald Test
Conduct a Wald test to compare the restricted AR(1) model against the unrestricted AR(2) model.
[h,p,Wstat,crit] = waldtest(r,R,V)
h = logical
0
p = 0.7166
Wstat = 0.1318
crit = 3.8415
The restricted AR(1) model is not rejected in favor of the AR(2) model (h = 0
).