Estimate p-values of fitted parameters using armax from sysid toolbox

1 view (last 30 days)
Hello!
I'm looking to find a way to find the p-value of the estimated parameters of an armax model fitted using the "armax" command of the System Identification Toolbox.
Any ideas?
Thank you

Accepted Answer

Ive J
Ive J on 1 Mar 2022
armax returns parameters and their uncertainties; then you can calculate the p-values.
load iddata9 z9
na = 4;
nc = 1;
sys = armax(z9,[na nc]);
[es, sd] = getpvec(sys);
p = (1 - normcdf(abs(es)./sd)).*2 % or 1 - chi2cdf((es./sd).^2, 1)
p = 5×1
0 0.4105 0.7223 0 0
Note that the assumption here is that estimates are from an MLE (i.e. assymptoticaly normal). See also here.

More Answers (0)

Categories

Find more on Linear Model Identification in Help Center and File Exchange

Products


Release

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!