1) `def QQPlot(cdf1, cdf2, ps):` ` res = []` ` for p in ps:` ` res.append((cdf1.Value(p), cdf2.Value(p)))` ` return res` 2) H: patient has tuberculosis C: patient has a cold E: patient goes to a doctor and reports a persistent cough My priors don't always have to be probabilities, as long as the ratio between them is right, P(H) ~ 1/15000 P(C) ~ 3 P(E|H) = 0.6 P(E|C) = 0.01 P(H|E) = (1/15000)(0.6) ------------------------- = 0.00133 or about one in 750 (1/15000)(0.6) + (3)(0.01) 3) Suppose that in a survey 19 out of 60 male respondents say they smoke, and 12 out of 40 female respondents say they smoke. 1) What is the probability that a randomly-chosen respondent is male? 60/100 2) What is the probability that a randomly-chosen respondent is a smoker? 31/100 3) What is the probability that a randomly-chosen respondent is a male smoker? 19/100 4) What is the probability that a randomly-chosen male is a smoker? 19/60 5) What is the probability that a randomly-chosen smoker is a male? 19/31 6) Does this mean that smoking is cool? Yes, smoking is cool. |

Lecture notes > Lecture 07 >