31 Lecture 02 - 2022

31.1 Bayesian data analysis

For each possible explanation of the data, count all the ways the data can happen. Explanations with more ways to produce the data are more plausible.

31.2 Misc

Area under the curve = 1 because it is normalized

1. No minimum sample size
2. Shape embodies samplen size
3. No point estimates since the distribution is the estimate (always use the entire distribution)
4. No one true interval

31.3 Using a posterior

A model’s behaviour is a function of all of its parameters at the same time, therefore you can’t just gaze at tables of parameters

• posterior predictive simulation
• model based forecasts
• causal effects
• counterfactuals
• prior predictions