Conditional probability and simulation
Probability given new information, Bayes' theorem for updating beliefs, and Monte Carlo simulation to check and estimate probabilities.
About 5 hours of study
Learning objectives
- Compute P(A given B) from counts and tables
- Update a probability with Bayes' theorem
- Estimate a probability by simulating many trials
Lessons
- 1
Conditional probability Intermediate
P(A given B): probability inside a smaller world, from a table of counts and from drawing cards without replacement.
- 2
Bayes' theorem Intermediate
Turning P(positive | sick) into P(sick | positive), and why a 99%-accurate test can still be wrong most of the time.
- 3
Simulation: estimating probabilities by experiment Intermediate
Monte Carlo simulation: repeating a random experiment many times to estimate a probability, the law of large numbers, seeds for repeatable runs, and how accurate an estimate is.
Module practice
- Conditional probability from a tableEasy
- Drawing balls without replacementMedium
- Questions about a two-way tableHard
- Total probability: defective partsEasy
- Is a positive test right?Medium
- Updating after each test resultHard
- Frequencies from recorded rollsEasy
- Estimating pi with random pointsMedium
- Simulating the Monty Hall gameHard