Answer to Assignment 2: 1. Here’s what we are given: P(disease) = 1/400 = 0.0025 P(no disease) = 399/400 = 0.9975 P( + | no disease) = 0.02 (these are false positives) P( – | disease) = 0.03 (these are false negatives) P( + | disease) = 0.97 (these are true positives) So we can apply Bayes’ Rule as follows: P( | disease ) P(disease ) P( | disease ) P(disease ) P( | nodisease ) P(nodisease ) (0.97)( 0.0025) P(disease | ) 0.108 (0.97)( 0.0025) (0.02)( 0.9975) P(disease | ) Or in words, the probability that a randomly selected person who got a positive test result actually has the disease is only about 10.8%. 2. Here’s what we are given: P(#1 has gold) = 1/3 P(#1 has goat) = 2/3 P(host shows #2 has goat | #1 has gold) = 1/2 (true regardless of Monty’s rule) P(host shows 2 has goat | #1 goat) = 1/4 (true if Monty always chooses which other door to open by flipping a coin; this is true because door #2 has a 1/2 chance of having the second goat and a 1/2 chance of being opened, so multiply these to get 1/4) So we can apply Bayes’ Rule as follows: Numerator = P(host shows #2 has goat | #1 has gold)∙P(#1 has gold) = (1/2)(1/3) = 1/6 Denominator = P(host shows #2 has goat | #1 has gold)∙P(#1 has gold) + P(host shows #2 has goat | #1 has goat)∙P(#1 has goat) = (1/2)(1/3) + (1/4)(2/3) = 1/3 P(#1 has gold | host shows #2 has goat) = (1/6) / (1/3) = 1/2 Thus, the chance the gold is behind door #1 is 1/2, and the chance it’s behind the other remaining door (#3) is also 1/2. Your odds of winning remain the same whether you switch or not. Importantly, this is true only if Monty follows the random choice rule, meaning there’s a chance he could have opened a door with gold. In the original description of the problem, he would only open a door with a goat.