Companion Resources to "Linear System Identification | System Identification, Part 2"

Companion Resources to "Linear System Identification | System Identification, Part 2"
Submitted by Brian Douglas on 11/01/2021
Reference 9 resources
Last Edited: 04/20/2022

These are the resources that are referenced throughout the MATLAB Tech Talk video I made called "Linear System Identification | System Identification, Part 2"

Here is the MATLAB Tech Talk video on linear system identification. If you've already seen the video and are just looking for the references that I used to make it then keep on scrolling! 

This video is a must watch if you want to understand the system identification workflow and how it can be realized using the tools in the System Identification Toolbox in MATLAB.

This article provides a nice gentle overview of the system identification process.  I reference it in the first video as well but I think it does such a nice job providing a visual overview that it's worth adding here as well.

This page is more like a collection of links than a description of linear model identification basics, however, all of the links are worth exploring if you'd like to understand about black box modeling, linear model structures, and more! 

This will provide a very brief overview of residual analysis. In the related topics at the bottom of the page you can find links to more information as well.

For a much fuller explanation of residual analysis, check out section 16.6 of this book.  Note, the link below goes to the publisher page for this book. 

I've posted the two simple MATLAB scripts to Github that I used in the video. I think a good way to use them would be to modify some of the parameters of the different system identification functions and see how it impacts the result. Check them out!

This is just the MATLAB command page for resid. I think it's worth digging in to if you'd like to see exactly how to create the residual plots.  Also, if you click on examples near the top you can see how to use this command in different situations.

Similarly, this is the MATLAB command page for goodOfFit. One thing to check out is the cost function section about halfway down the page. There it shows the formula for the normalized root mean squared error calculation that is used in the "compare" function (and what I mention in the video).