This preview shows pages 1–10. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: ECON 4109H LECTURE 8 Expected Utility Theory October 6, 2010 ECON 4109H LECTURE 8 Expected Utility Theory Finite set X = { 1, 2, . . . , n } . lottery = a probability distribution over X . But what is a probability distribution? ECON 4109H LECTURE 8 Expected Utility Theory lottery = a vector ( p 1 , p 2 , . . . , p n ) p k = probability of k . Assume: 1 p k > 0 for all k in X . 2 ∑ n k = 1 p k = 1. Assumptions imply p k 6 1 for all k . ECON 4109H LECTURE 8 Expected Utility Theory Δ = set of all lotteries on X . Notation: δ = ( p 1 , . . . , p n ) δ = ( p 1 , . . . , p n ) δ 00 = ( p 00 1 , . . . , p 00 n ) ECON 4109H LECTURE 8 Expected Utility Theory For lotteries δ = ( p 1 , . . . , p n ) and δ = ( p 1 , . . . , p n ) and α ∈ [ 0, 1 ] : Form a compound lottery αδ + ( 1 α ) δ . αδ +( 1 α ) δ means: with probability α , δ happens and with probability 1 α , δ happens. Formally: αδ + ( 1 α ) δ = ( α p 1 + ( 1 α ) p 1 , . . . , α p n + ( 1 α ) p n ) ECON 4109H LECTURE 8 Expected Utility Theory Theorem If δ and δ are lotteries, then so is αδ +( 1 α ) δ for any α ∈ [ 0, 1 ] . ECON 4109H LECTURE 8 Expected Utility Theory Theorem If δ and δ are lotteries, then so is αδ +( 1 α ) δ for any α ∈ [ 0, 1 ] . We must verify the properties of a probability distribution: 1 p 00 k > 0 for all k in X . 2 ∑ n k = 1 p 00 k = 1. ECON 4109H LECTURE 8 Expected Utility Theory Theorem If δ and δ are lotteries, then so is αδ +( 1 α ) δ for any α ∈ [ 0, 1 ] . α p k + ( 1 α ) p k > This verifies the first property. n X k = 1 [ α p k +( 1 α ) p k ] = α n X k = 1 p k +( 1 α ) n X k = 1 p k = α +( 1 α ) = 1 This verifies the second property. ECON 4109H LECTURE 8 Axioms for Expected Utility Theory Completeness Either δ 4 δ or δ 4 δ ....
View
Full
Document
This note was uploaded on 01/19/2012 for the course ECON 4109H taught by Professor Notsure during the Fall '09 term at Minnesota.
 Fall '09
 Notsure
 Utility

Click to edit the document details