I've read that multicollinearity can lead to a linear regression model being unstable. I came across this wikipedia article: https://en.wikipedia.org/wiki/Stability_(learning_theory), that basically says a model is unstable if the outputs vary wildly for small changes to the input. I was wondering if there is a more precise way of defining this.
I have also seen this question posted earlier (Model stability and variability), but I'm still not certain I fully understand what we are looking for when we are trying to test for stability.
