The DAG gets informed by the experiments that you already have and basically people's intuition for the business. So if I want to get that in, I'm going to say that prices cause your likelihood to use our product. That's a statement about causality. And this is a statement about a part of a DAG. If someone else came back to me and they said, actually, your use of the ride causes the price. That would be a different DAG. But yeah, so the DAG rules out certain models that you'd like to be able to use. Those feedback loops are just really, really hard to estimate.
What causes us to keep returning to the topic of causal inference on this show? DAG if we know! Whether or not you're familiar with directed acyclic graphs (or… DAGs) in the context of causal inference, this episode is likely for you! DJ Rich, a data scientist at Lyft, joined us to discuss causality — why it matters, why it's tricky, and what happens when you tackle causally modelling the complexity of a large-scale, two-sided market! For complete show notes, including links to items mentioned in this episode and a transcript of the show, visit the show page.