bayes-rules-notes/R/misc/beta-binomial.qmd

45 lines
1.6 KiB
Plaintext

---
title: "Beta-Binomial Bayesian Model Example"
author: "Emanuel Rodriguez"
execute:
message: false
warning: false
format:
html:
monofont: "Cascadia Mono"
highlight-style: gruvbox-dark
css: styles.css
callout-icon: false
callout-apperance: simple
toc: false
html-math-method: katex
---
## Introduction
In this short article I will try to put together a set of notes to
go over the Beta-Binomial Bayesian Model. The example used here is pretty
trivial but so is this model so :shrug:
## When to use the Beta-Binomial?
The situations it makes sense to use the Beta-Binomial Model (BBM) are
cases where we have some conditional dependence between a random
variable $Y$ and a continuous parameter $\pi \in [0, 1]$. Where $Y$ is the number of successes in $n$ independent trials, each with a probability of
success $\pi$.
For the sake of these notes, let us start by setting up an example that will
allow us to illustrate the implementation of this model. Suppose there
is a slot machine manufacturer who recently was accused of embedding software
into the hardware that would cause more losses than usual to players.
To be allowed to continue selling machines to casinos, they must go through a
series of inspections. The body governing the sale of these machines has tasked
us to determine whether or not the software has been completely removed or if,
instead, the company refined the bias to make it less noticeable.
## What we already know
The set of facts that we can put together to build a known distribution