Inferring the effect of an event using CausalImpact by Kay Brodersen
Why Cross Validated Stack Exchange recommends it
Cross Validated users link it when people ask how to measure before-and-after effects of an event in a time series or how Bayesians infer causal effects from observational data.
Recommended by 3 people across 3 different months, 2018 to 2023.
What it is
A 30-minute conference talk by Kay Brodersen of Google at Big Data Spain 2016 introducing the CausalImpact method and R package. It shows how to use Bayesian structural time-series models to estimate the causal effect of an intervention when a randomized experiment is not available.
What people said
Recommends Mark White · 2018-04 · Cross Validated Stack Exchange · on “How to statistically measure the before and after effects of a change in a time series”
check out the CausalImpact R package: [LINK] with an introduction video here: [this video] I don't like all the fake or perfect data circumstances in their examples, so here's something I did with it: [LINK]
Recommends greenglas · 2022-01 · Cross Validated Stack Exchange · on “Can we talk about statistical significance using Bayesian Inference?”
@christianHennig You are right, it is not strictly developed by Google. However it is the one recommended and used by Google. A very helpful video is the following: [this video] where KH Brodersen (Senior Quantitative Analyst at Google) goes through the package.
Recommends Durden · 2023-05 · Cross Validated Stack Exchange · on “Bayesian Experiment Design Resources”
It's somewhat ambiguous how you phrased the question. If you're interested in how Bayesian think about _trial design_ I recommend starting with this presentation by Frank Harrell [LINK] . But if you're interested in how Bayesians use _observational data_ to infer causal effects otherwise needing RCTs, start with this talk by Kay Brodersen [this video] .