60 likes | 175 Views
Value of information. Objective: Quantify value of information --both perfect and imperfect Principle: Information should help a decision maker make decisions that are better than decisions without information
E N D
Value of information • Objective: Quantify value of information --both perfect and imperfect • Principle: Information should help a decision maker make decisions that are better than decisions without information • Expected value of information=expected monetary value (EMV) with information-expected monetary value (EMV) without information
Measuring accuracy of information • Value of information depends on decision we want to make and accuracy of information • Accuracy of information: P(expert says “A”/A occurs), P(expert says “A will not occur”/A does not occur). The higher these conditional probabilities, the more accurate is the information. • Perfect information: P(expert says “A”/A occurs)=1, P(expert says “no A”/A does not occur)=1.
Example I (0.1) 20 (0.2) 10 EMV=7 (0.6) 5 A (0.1) 0 B 6 If we have no information we will chose A; EMV with no information=7
Example: continued (0.1) A20 Choose A, payoff 20 (0.2) A 10 Choose A, payoff 10 A 5 (0.6) Choose B, payoff 6 A 0 (0.1) Choose B, payoff 6 EMV with perfect information=8.2 Value of perfect information=8.2-7=1.2
Example II (0.1) 20 (0.2) 10 (0.6) 8 A (0.1) 7 B 6 Option A dominates option B. Value of perfect information =0.