Hi Crawbears,
In a previous post, we applied chained Bayesian logic to interpret a positive COVID-19 test result. We’ve also constructed a couple of simulations, for example, here and here.
Today, we’ll construct a Bayesian simulation to demonstrate how a Bayesian update based on new observations improves our prior belief about the world, in this case about systolic blood pressure. We’ll start with our “prior” and, after observing new information, run simulations that update our understanding with a posterior distribution that gets closer to the observed distribution with each new observation. Let’s dive in.
Pretty cool seeing Bayesian updates in action, huh? Next time, we’ll fit linear and logistic models to housing prices and child autism data sets.
Keep sending your questions to info@crawstat.com, we love hearing from you!
Till next time,
Rish
