02_Decision Making under Risk Flashcards

1
Q

2 Main Methods for Decision Making under Risk

A
  • Expected Utility Theory
  • mu-sigma Rule
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Decision Situtation
under Risk

5 items

A
  • multiiple scenarios with probabilities
  • single goal
  • single DM
  • static decision context
  • game against nature
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Certainty Equivalent
CE(L)

A
  • Certainty Equivalent of a lottery L is the # CE(L) such that the DM is indifferent between the lottery and receivin a certain payement of CE(L)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Risk Premium
RP(L)

A
  • the Risk premium of a Lottery L is the difference between the expected value of the lottery EV(L) and the Certainty equivalent of the lottery CE(L)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

3 main types of utility functions

A

- linear
- convex (x^2)
- increase of utility is positive and increasing with x
- concave (x^0.5)
- increase of utility is positive and decreasing with x

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Arrow-Pratt Measure

Definition

A
  • reveals the risk attitude of the DM at x
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

3 different type of DMs

A
  • risk averse
  • risk seeking
  • risk neutral
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Utility Function of
Risk Averse DM

A
  • concave function
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Utility Function of
Risk Seeking DM

A
  • convex function
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Utility Function of
Risk Neutral DM

A
  • linear function
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What did Friedman and Savage (1948) observe regarding utility functions+

A
  • utility function is mixed rather than only one type of utility function and depends on specific situation
  • Situation I: Concave
  • Situation II: Convex
  • Situation III: Concave
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Standard deviation

mu-sigma rule

A
  • sqrt of Variance
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Variance of the outcome of action a(i)

mu-sigma rule

A

[pj * e(i,j)^2] - EV(ai)^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

mu(ai) / EV(ai)

mu-sigma rule

A

Sum of (e(i,j) * pj)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Expected Utility
vs.
mu-sigma Rule

When will both rules lead to same decision?

A

in case of:
- risk neutral decision maker
- quadratic utility function and appropriate calibration of preference function
- normally distributed outcomes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Drawbacks of mu-sigma rule

A
  • does not guide the DM in calibrating the preference function
  • cannot determine the risk attitude of the decision maker
17
Q

Decision Making under Risk
Criteria

5 items

A
  • risk
  • one criterion
  • single DM
  • static decision
18
Q

Arrow-Pratt Measure (AP)
Decision Rule regarding Risk Attitude

A
  1. Risk Neutral
    AP(x) = 0
  2. Risk Averse
    AP(x) > 0

3.Risk Seeking
AP(x) < 0

19
Q

Arrow-Pratt Measure
Formula

A

AP(x) = - [u’‘(x) / u’(x)]

20
Q

Risk Averse Attitude
- form of utility function
- CE(L) to EV
- RP(L)
- AP

A
  • concave utility function
  • CE < EV
  • RP > 0
  • AP > 0 as u’‘(x) < 0
21
Q

Risk Seeking Behaviour
- form of utility function
- CE(L)
- RP(L)
- AP

A
  • convex utility function
  • CE(L) > EV
  • RP < 0
  • AP < 0 as u’‘(x) > 0
22
Q

When can the expected value criterion be used for solving a decision problem? Characterize the decision context.

A
  • risk
  • single criterion
  • Single risk neutral DM
  • static
  • randomly behavin opponent
23
Q

Which axioms must be satisfied for a DM to use expected utility criterion?

A
  1. Complete Order
  2. Continuity
  3. Independence
  4. Compound Lotteries
  5. Unequal Probabilities