I'm still new with the system identification subject but I have a gist of it. I had obtained the experiment data and are using MATLAB's ARMAX function to build ARIMAX model structure. I'm trying the black box approach and I was wondering the combinations of order I should choose. I tried using MATLAB's AIC and tried to find the lowest AIC. However, when I choose the model with the lowest AIC and compared it, it turns out bad and return with a huger error based on the graph.
Hi, I would say the model order simply depends on what you want to identify.
If the behaviour match better with first order system, then use first order system and etc.
Regards
Reza
Hi Amelia,
The System Identification Toolbox for Matlab (i assume that is the program that you use for the models ) it's based on the book "System Identification:Theory for the user" by Lennart Ljung and explain all the different models (ARX,ARMA,ARIMAX,etc) , in this URL: https://la.mathworks.com/videos/lennart-ljung-on-system-identification-toolbox-advice-for-beginners-96988.html
The autor explain the basics about this toolbox, I hope this message has helped you
Hi, thank you for all the replies.
Reza Dariani, I'm actually trying a black box approach, that's why I can't really decide on a model order.
Michael Short , I have seen the help page for AIC and I followed the right procedures in using the AIC. The experiment data actually was preprocessed before I use the ARIMAX. But, the result is still the same, the minimum AIC didn't produce a good model when I tried to compare it with the original data.
Camilo Vicente Gonzalez Pozo, I don't think the toolbox uses AIC, and it automatically chooses the best model. My MATLAB doesn't seem to have ARMAX or ARIMAX in the toolbox, so I use the code instead.
- Badges
- Science topic
- Similar topics
- Condensed Matter Physics
- Zero