Informative vs non-informative priors — the tradeoff.

medium

Answer

$Non-informative (flat, Jeffreys): 'let the data speak' — safe when you have plenty of data, but can be worse than an informative prior at small n.$
$Informative (based on historical data, domain knowledge, similar experiments): dramatically improves inference in small-sample settings, but can also inject bias if wrong.$
$Rule: (1) when n is large, prior barely matters.$
$(2) When n is small, informative priors are your friend if you have credible data.$

Check yourself — multiple choice

Same thing
Non-informative: safe at large n; informative: helpful at small n if credible, harmful if wrong; prior barely matters at large n
Only non-informative
Random

Non-informative: safe at large n; informative: helpful at small n if credible.

#bayesian

215 interview questions in this topic.