I’m going to be attempting to re-enter writing about econometrics. I’ve just a few papers I must work by means of intently, and it’s only a good time to maintain retooling extra typically. I additionally need to give my subscribing readers extra worth for being a subscriber, and due to this fact am simply going to return to my randomized paywalls. And immediately, I had Cosmos as soon as once more flip a coin 3 times to see if I’d be paywalling and certainly, immediately is a randomized paywall day!
However earlier than I do, I wished to inform you a bit of about what this put up is about. This put up goes to be half 1 in a few posts that exhibits the precise circumstances below which you’ll establish utilizing diff-in-diff each an unweighted causal impact and a inhabitants weighted causal impact. I’m going to point out you this formally utilizing one thing I’ve been engaged on. After which I’m going to point out you in a subsequent put up code and output from a simulation that illustrates it. However the punch line is that this:
simply because parallel traits holds with mixture knowledge doesn’t imply it can maintain with inhabitants weighted knowledge. The circumstances below which each maintain and never maintain is what this collection is about.
So with that, let’s go! Thanks once more for all of your help! Please take into account changing into a paying subscriber. Provide curves slope upwards and after a protracted hiatus of paywalling, I’m going to be going again to doing so and writing about econometrics, and new papers I’m investing time in, notably as I put together for my new programs subsequent spring at Harvard (together with a PhD course).
