This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 Decision Making 2 Cornel University Announcements Please help me learn names by sitting in the same part of the room Read the syllabus Problem Set #1 will be posted tomorrow Problem sets are optional, not graded, but highly recommended Answers will be posted Friday Am I too fast/slow, talk too loud/soft? Let me know! 3 Cornel University Results from Quiz Question 1: 18 out of 55 correct Question 2: 18 out of 55 correct Question 3: 0 out of 55 correct Question 4: 2 out of 55 correct Our goal: lets get to 55 out of 55 on all questions by October 16 4 Cornel University Decision Making Goals After todays class, you should be able to: Evaluate business decisions in a world of uncertainty Calculate the worth of information Understand and value real options 2 5 Cornel University Decision Trees Simple tool to organize decision process Trees run left to right chronologically Square = Decision nodes Circle = Chance nodes Outcome branches are lines Win Lose 0.5 0.5 Dont buy Buy $0 $-1 $9 (10-1) Buy a $1 lottery ticket w/ a $10 payout? 6 Cornel University Decision Trees (cont) Payoff at the end of a branch Chance node probabilities on outcome branch Chance node probabilities must sum to one (outcomes mutually exclusive) Expected value (EV) of chance node: EV = (prob1 * outcome1) + (prob2 * outcome2) Express all outcomes in NPV terms Win Lose 0.5 0.5 Dont buy Buy $0 $-1 $9 (10-1) Buy a $1 lottery ticket w/ a $10 payout? 7 Cornel University Backward Induction Maximizes EV For each of the right-most nodes: If a decision node ( ), determine best alternative to take. The payoff from this alternative is value of the decision. If a chance node ( ), calculate the expected value. This is the value of the chance mode. For the nodes one step to the left: If a decision mode, determine the best alternative to take using future nodes If a chance node, calculate the expected value using the value of future nodes as payoffs. Repeat until left-mode node is reached crossing off sub- optimal decisions. 8 Cornel University Old MacDonald Problem Old MacDonald has a farm. On her farm she can grow either tomatoes or cacti. Cacti are drought resistant, while tomatoes are not. Cacti will produce a profit of $18,000 regardless of the weather, while a crop of tomatoes will produce a profit of $30,000 if there is no drought and $10,000 if there is a drought. The probability of a drought is 1/2....
View Full Document
This note was uploaded on 02/20/2009 for the course AEM 4240 taught by Professor Blalock,g. during the Fall '07 term at Cornell University (Engineering School).
- Fall '07