From the course: Machine Learning & AI Foundations: Linear Regression

Unlock the full course today

Join today to access over 24,800 courses taught by industry experts.

SEM

SEM

- [Instructor] Okay, let's briefly talk about a whole family of techniques called path analysis and/or structural equation modeling. The two phrases are actually different in subtle ways. Now this is a big topic. Entire books have been dedicated to it, but I want to give you a feel for how this addresses issues that although somewhat possible in regression, become extremely difficult. One way to think of it is it somewhat combines regression and factor analysis at the same time. Let's take a look at about the easiest example that we can to get a feel for this. How would you try to prove that A predicts B, which in turn predicts C? Now, there's endless conversation that you can find in academic papers and books and on the web about mediation and moderation. And if you're guessing that it's similar to this, it is related. But let's just stick to how structured equation modeling would tackle this, starting with how you could try to get this with regression. First, you'd have to establish…

Contents