Lecture 1 theoretical concepts

Question 1

Name three decision types

Answer 1

1. Decision making under certainty
2. Decision making under uncertainty (know nothing about probability outcomes
3. Decision making under risk: know probability distribution outcomes

Question 2

Name Different decision making criteria

Answer 2

1. Maximax criteria
2. Maximin criteria
3. Hurwicz criterion
4. Laplace criterion
5. Minimax regret criterion

Question 3

What is maximax criterion?

Answer 3

You have to determine which outcome is the best for each decision and choose the decision with the highest outcome.

Question 4

What is the maximin criterion?

Answer 4

You have to determine the worst outcome for each decision and choose the decision where the worst outcome is the highest

Question 5

What is the Hurwicz criterion?

Answer 5

This is a weighted average between optimism and pessimism. You have to select a coefficient between 0 and 1, where 1 is optimistic, and then calculate weighted average for each alternative decision and select the alternative with the highest value

Question 6

What is the minimax regret criterion?

Answer 6

Regret = the difference between actual outcome and the best possible outcome. You have to determine the maximum regret per alternative. Then select the alternative where the maximum regret is the lowest.

Question 7

What is the outcome bias?

Answer 7

This is a bias where decision makers only evaluate decisions based on outcome and not on process

Question 8

What makes for a good decision?

Answer 8

1. It evaluates the outcome
2. It evaluates the processes

Question 9

How do you make decisions under risk?

Answer 9

When the probability distribution is known, you want to maximize the expected monetary value

Question 10

What is expected monetary value?

Answer 10

This is the weighted average of expected profits

Question 11

What is meant by the expected value with perfect information (EVwPI)?

Answer 11

This is the average long-term outcome when having perfect information

Question 12

What is meant by the Expected Value Perfect Information (EVPI)?

Answer 12

This is the value of perfect prior knowledge and is calculated as EVwPI - Maximum EMV

It is always the same to minimum expected opportunity loss

Question 13

What is expected opportunity loss?

Answer 13

This is the weighted average of expected regret

Question 14

What is sensitivity analysis?

Answer 14

This analysis concerns itself with the question: will decisions change when the input variables change? Often, the specific probabilities are replaced with P and (1-P).

Question 15

What is a robust model?

Answer 15

A robust model means that a decision doesn’t change with highly-varied input parameters