## Metadata - Author: [[itamar-faran|Itamar Faran]] - Full Title:: How to Do Bayesian a/B Testing at Scale - Category:: #🗞️Articles - URL:: https://towardsdatascience.com/how-to-do-bayesian-a-b-testing-fast-41ee00d55be8 - Finished date:: [[2024-05-06]] ## Highlights > The [Beta-Binomial](https://en.wikipedia.org/wiki/Beta-binomial_distribution) model is used to model binary data such as conversions or clicks (“did the user convert or not?”). I’ll also review the [Normal-Normal](https://en.wikipedia.org/wiki/Multivariate_normal_distribution#Bivariate_case_2) model, which is used for continuous data (for example revenue per user). > In the Beta-Binomial model we assume that the conversion rate *Pᴀ* has a [Beta distribution](https://en.wikipedia.org/wiki/Beta_distribution) with parameters *α* and *β*. A [common choice](https://en.wikipedia.org/wiki/Beta_distribution#Bayes-Laplace_prior_probability_(Beta(1,1))) for *α* and *β* is 1, which results with a uniform distribution (sometimes referred to as an “[uninformative prior](https://en.wikipedia.org/wiki/Prior_probability#Uninformative_priors)”). I discuss the choice of the prior in more detail in the appendix, but for now let’s continue as if they have already been chosen. ([View Highlight](https://read.readwise.io/read/01hx3cw5w7we9mcsnkbbmrzjrk))