{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# Lect16 - Course Logisics We need more TAs Or less work for...

This preview shows pages 1–7. Sign up to view the full content.

Course Logisics We need more TAs Or less work for Dae Hoon Grading is the time- consuming portion… Machine Problem?

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Religious wars among Statisticians (are you a Bayesian?) Frequentist / objectivist / classical statistics / Fisherian World is some true distribution Data are a random sampling We can come to know the World approximately via data Hypothesize; Observe; Evaluate Changing the hypothesis taints the data (baseball, lottery) Stock scam, wrong hypothesis Bayesian Evidence can be objective (data) or subjective Evidence can testify for / against different distributions Will my plane crash? Chance of rain? Two meter problem Two envelope problem
Probability / Statistics Probability Space Sample space (Atomic Events) Event space Probability measure Random Variables Distributions Statistical Inference

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Joint Probability Distribution Discrete random variables Encodes all information of interest Allows arbitrary dependencies Exhaustive and Exclusive Atomic Events Must sum to 1 (one less degrees of freedom than entries) Diseases / Symptoms illustration (useful but overly specific; really evidence and conclusions)
Sunny Cloudy Rainy Snowy Yes 0.25 0.15 0.05 0.13 No 0.05 0.1 0.25 0.02 Weather Have Fun? P(Rainy) = 0.05 + 0.25 = 0.3 P(Rainy | Sunny) = P(Rainy   Sunny) P( Sunny) = 0.3 / 0.7 0.43 P(Fun | Sunny) = P(Fun Sunny) P(Sunny) = 0.25 / 0.3 0.83 Do I prefer Sun or Snow? P(Fun | Snowy) = 0.13 / 0.15 0.87 So I prefer snow

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Bayes Theorem P(A | B) = Easy to rederive (you should never get it wrong!) P(A B) = P(A | B) P(B) = P(B | A) P(A) Equate and solve for P(A | B) P(B | A) P(A) P(B)
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}