Updating Beliefs with New Evidence
What You'll Learn
You'll learn how to use Bayes' Theorem to revise your probability estimates when you receive new information.
The Core Idea
Imagine you're a doctor who initially believes a patient has a 5% chance of having a rare disease (your prior probability). Then a test comes back positive. That's new evidence. Bayes' Theorem tells you exactly how to update that 5% to a new, more accurate probability (your posterior probability) based on the test result.
This is the power of Bayesian updating: systematically incorporating new data to refine your beliefs.
How It Works
Recall Bayes' Theorem from the previous lesson:
P(A|B) = [P(B|A) × P(A)] / P(B)
In the updating framework:
- P(A) = your prior belief (what you thought before seeing evidence)
- B = the new evidence you just observed
- P(A|B) = your posterior belief (updated probability after evidence)
The formula tells you how much weight to give the evidence based on:
- How likely the evidence would appear if A were true: P(B|A)
- How likely the evidence is overall: P(B)
A Simple Example
You think there's a 30% chance it will rain today (prior). Then you see dark clouds (evidence). If rain usually comes with dark clouds 80% of the time, but dark clouds appear only 40% of days overall, Bayes' Theorem calculates your updated rain probability:
P(Rain|Clouds) = (0.80 × 0.30) / 0.40 = 0.60
Your belief jumps from 30% to 60% after seeing the clouds.
Key Takeaway: Bayes' Theorem provides a mathematical framework for updating probabilities as new evidence arrives, transforming prior beliefs into posterior beliefs in a principled, quantitative way.