week6da - Spring, 2010 Spring, DecisionAnalysis MS405 Value...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Spring, 2010 Spring, 2010 Week 6 1 Decision Analysis   MS 405    Value of Information Kemal Kılıç Faculty of Engineering and Natural Sciences
Background image of page 1

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

View Full DocumentRight Arrow Icon
Spring, 2010 Spring, 2010 Introduction Week 6 2
Background image of page 2
Spring, 2010 Spring, 2010 Week 6 3 Introduction
Background image of page 3

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

View Full DocumentRight Arrow Icon
Spring, 2010 Spring, 2010 Information vs. Uncertainty Basic Principle: Information only has value in a decision problem if it results in a change in some action to be taken by a DM. Information vs. Uncertainty Uncertainty is the lack of knowledge captured in probability distributions. New information may affect our probability distributions through a Bayesian Revision. Effect of new information is to reduce the amount of uncertainty. Information gathering is a normal activity in the decision making process. Week 6 4
Background image of page 4
Spring, 2010 Spring, 2010 Sources of Information Observation/empirical data: test results, results of surveys, experiments, samples, etc. Experts’ opinions: forecasts, estimates, predictions, etc. Waiting: the best way to determine what will happen with next year’s sales is just to wait and find out. Benefits and Costs of Information Presumably, more information allows us to make better decisions. Some important questions . . . Are the benefits to be expected worth the costs? How much information should we buy before we make a decision? Is a particular source of information worthwhile? Week 6 5 Information vs. Uncertainty
Background image of page 5

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

View Full DocumentRight Arrow Icon
Spring, 2010 Spring, 2010 Information is seldom perfectly reliable. Generally, information does not totally eliminate uncertainty. experts are fallible forecasts are inaccurate test results may be wrong Def.: “Perfect information” is information that is perfectly reliable, it predicts outcomes correctly 100% of the time and therefore, resolves uncertainty. Its like a “crystal ball”. (Clairvoyance) Perfect Information vs. Imperfect Information Week 6 6
Background image of page 6
Spring, 2010 Spring, 2010 Expected value of perfect information Perfect information rarely exists, but is a useful concept because: It’s easy to calculate the expected value of perfect information. It provides an upper bound on the value of real information since it is a “best case” scenario. It answers the question”how much better off would I be right now if I could make the decision after knowing what outcome will occur?” Week 6 7
Background image of page 7

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

View Full DocumentRight Arrow Icon
Spring, 2010 Spring, 2010 Perfect Information Consider the coin flip example A1: Flip a fair coin: H - Win $10 T - Lose $2 A2: Flip a fair coin H - Win $2 T - Lose $1 Suppose a friend says he is a clairvoyant, and can predict which event will occur before it occurs, but he cannot make an event occur . Expected Value of the lottery without perfect information is
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/19/2011 for the course ECON Econ203 taught by Professor Majdabpadawnan during the Spring '11 term at University of Maribor.

Page1 / 46

week6da - Spring, 2010 Spring, DecisionAnalysis MS405 Value...

This preview shows document pages 1 - 9. Sign up to view the full document.

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