bayessox - Bayesian Inference Inference Regarding a...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Bayesian Inference – Inference Regarding a Proportion Probability Red Sox Beat Yankees in a Game Parameter: π = Probability that Boston would beat NY in a baseball game (0 π 1) Prior (to season) belief regarding the value of π : No information (anywhere between 0 and 1, equally likely) Values closer to ½ are most likely with no “bias” away from ½. “Expect” that π ≈ 2/3. Family of “prior distributions” with nice properties and very flexible: Beta Distribution: Support is from 0 to 1, indexed by 2 parameters: α,β Density Function: ) 1 ( ) ( ) ( ) ( )! 1 ( ) ( 0 , , 1 0 ) 1 ( ) ( ) ( ) ( ) ( 2 1 1 + + + = + = - = Γ - Γ Γ + Γ = - - β α αβ π V E n n p Case 1: Non-informative Prior ( α = β = 1 29 p( π )=1 0 2 y 0 .2 .4 .6 .8 1 x
Background image of page 1

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

View Full DocumentRight Arrow Icon
Case 2: Informative Prior with mean 0.5 ( α = β = 2 29 p( π )=3 π(1-π29 Case 3: Informative Prior with mean 0.67 ( α = 2 , β = 1 29 p( π )=2 π 0 .5 1 1.5 y 0 .2 .4 .6 .8 1 x 2 0 .2 .4 .6 .8 1 x
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 4

bayessox - Bayesian Inference Inference Regarding a...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online