Can I get volatilities out of the GARCH11 model?

Hello everyone! New to this site (and noob in bayesian stats).

I’ve been following this package for quite a bit of time and I was looking for an excuse to give it a try. I recently saw that at my new job we use a lot of garch models to get volatilities over a sample with daily frequency. I thought it would be cool (and useful) to use pymc3 to get a full distribution of the volatility for any given date, but the parameters are omega, alpha1, beta1, not the sigma itself, so Idk how to recover it. I did see that there was some discussion over github about this (can’t find it now :sweat_smile:) but it didn’t go anywhere. It’s weird because garch models are typically used to get the volatility.

Is there any way to do this? Is the garch model dev discontinued?

Thank you!


Could you post the code or NB you used to model this?

Sorry I didn’t have notifications turned on.

After writing a reply I saw that I can recover the volatilities manually anyway, just need to do it iteratively with all the alphas and betas, my bad!