Causality and time series – when an econometrician goes Bayesian
In this section, we’re going to introduce a new style of thinking about causality.
We’ll start this section with a brief overview of quasi-experimental methods. Next, we’ll take a closer look at one of these methods – the synthetic control estimator. We’ll implement the synthetic control estimator using an open source package, CausalPy, from PyMC Labs and test it on real-life data.
Quasi-experiments
Randomized controlled trials (RCTs) are often considered the “gold standard” for causal inference. One of the challenges regarding RCTs is that we cannot carry them out in certain scenarios.
On the other hand, there’s a broad class of circumstances where we can observe naturally occurring interventions that we cannot control or randomize. Something naturally changes in the world, and we are interested in understanding the impact of such an event on...
