Scribd
Upload
19 views

Week 2 Lecture

You're correct that the value function in prospect theory plays a similar role to the utility function in expected utility theory. The key differences are: 1) The value function is defined over gains and losses relative to a reference point, rather than final outcomes. It allows for loss aversion. 2) The weighting function transforms probabilities in a non-linear way, capturing probability weighting effects. So prospect theory provides a better descriptive model of choice under risk by incorporating reference dependence and non-linear probability weighting, even though the value function serves a similar role to utility in representing preferences over outcomes. The terms are used for convenience but do capture important psychological differences from EUT.

Uploaded by

Mariana Ramos
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
19 views

Week 2 Lecture

You're correct that the value function in prospect theory plays a similar role to the utility function in expected utility theory. The key differences are: 1) The value function is defined over gains and losses relative to a reference point, rather than final outcomes. It allows for loss aversion. 2) The weighting function transforms probabilities in a non-linear way, capturing probability weighting effects. So prospect theory provides a better descriptive model of choice under risk by incorporating reference dependence and non-linear probability weighting, even though the value function serves a similar role to utility in representing preferences over outcomes. The terms are used for convenience but do capture important psychological differences from EUT.

Uploaded by

Mariana Ramos
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 23

EC345|Behavioural Economics:

Theory and Applications


Week 2: Decision Under Uncertainty
Dr Mahnaz Nazneen
The Expected Utility Theory

Anomalies/ Puzzles
Week 2
Recap A Behavioural Theory- Prospect Theory

Empirical Applications
Example
Choice 1 Choice 2
– option A offers a 33% chance at £2500 and a 67% – option C offers a 33% chance at £2500, a 66%
chance at £0; chance at £2400, and a 1% chance at £0;
– option B offers a 34% chance at £2400 and a 66% – option D offers £2400 for sure.
chance at £0.

• Most people prefer A to B; also, choose D over C


• Choice 2 is same as Choice 1:
• add a common consequence to both A and B to get C and D: a 66% chance at
$2400
• Violation of independence axiom
• This effect is called certainty effect [also known as common consequence effect]
Prospect Theory- Value Function
Three main principles:
• Reference dependence: outcomes are
viewed relative to a reference point and hence
coded as gains or losses
• Diminishing sensitivity: changes in value
have a greater impact near the reference
point than away from the reference point
• Loss aversion: pain from losses has
stronger effect than pleasure from gains
Prospect Theory-
Value Function
Note the shape of value
function implies:
• concave in gains domain
(risk averse)
• Convex in losses domain
(risk seeking)
• Steeper for losses than
gains
Prospect Theory-
Weighting Function
Two main principles:
• Reference dependence- probabilities are
weighed in reference to two reference points –
impossibility (0%) and certainty (100%)
• Diminishing sensitivity- People are most
sensitive to changes in probability when they are
near 0% or 100%, than when the change applies to
intermediate probabilities.
• π(p) is concave for small probabilities and
convex for medium and large probabilities
• People overweigh small probabilities and
underweigh medium to large probabilities. Fig: A typical π(p) function, which is concave for small
probabilities and convex for medium to large probabilities
Application- Mental Accounting
• Examples of how individuals and households make consumer and
financial decisions based on subjective valuation of money
• treat money as less fungible
• Mental accounting suggests that people prefer to
• segregate gains [e.g. £100 from parents and £50 from lottery- most
people find it more enjoyable to view these outcomes separately,
£100 plus £50, than together, £150.]
• aggregate losses [a £100 speeding ticket and a £50 online bank scam-
a loss of £150 seems less painful than two separate losses of £100 and
£50]
Tanaka, Camerer, & Nguyen (2010)
• Evidence from developed countries suggest wealthier people are willing to take
more risk than poorer people
• Evidence from developing countries is more ambiguous
• Measure risk and time preference in the field combining both survey and
experiment data in Vietnam
• designed to collect preference measures experimentally (2005) and correlate
those measures with demographic and economic variables (income, in particular)
from the previous household survey (2002)
Tanaka, Camerer, & Nguyen (2010)
Tanaka, Camerer, & Nguyen (2010)
Measure parameters of prospect theory (σ, λ and α), rather than
just concavity of utility function in EU
• Value Function
𝑣 𝑥 = 𝑥 𝜎 for x > 0
𝑣 𝑥 = −𝜆(−𝑥 𝜎 ) for x < 0
• Weighting Function

1
𝜋 𝑝 =
1
exp[ln(𝑝)𝛼 ]

• Parameters σ, λ represent concavity of the value function, and the degree of loss aversion
Tanaka, Camerer, & Nguyen (2010)
Tanaka, Camerer, & Nguyen (2010)
Tanaka, Camerer, & Nguyen (2010)
Tanaka, Camerer, & Nguyen (2010)
• not many subjects whose choices are consistent with EU
• Mean estimated values of (σ, α) are (0.59, 0.74) and (0.63,0.74)
• mean value of α is significantly different from one at the 1% significance level- rejecting EU in
favour of inverted-S shaped probability weighting (as suggested by Kahneman and Tversky)
• average estimated value of loss aversion (λ) is 2.63 (Kahneman and Tversky
estimated it to be 2.25)
• Next- regress the curvature of utility function (σ) using OLS and loss aversion (λ)
by Maximum Likelihood against individual variables
Results:
• subjects who are more
educated and older are more
risk averse.
• ethnic Chinese are less loss
averse
• people living in the south are
more loss averse
individuals living in wealthier
villages are less loss averse
and also less risk averse.
Tanaka, Camerer, & Nguyen (2010)

important result: mean village


income is related to risk preferences

People living in poor villages are not


necessarily afraid of uncertainty, in
the sense of income variation;
instead, they are averse to loss.
Disposition Effect:
tendency of
investors to hold
losing investments
too long and sell
winning
investments too
soon

11930 13883
𝑃𝐿𝑅 = = 0.098 𝑃𝐺𝑅 = = 0.148
11930 + 110348 13883 + 79658
Disposition
Effect
Disposition Effect (Contd.)
• Why is the disposition effect usually not observed
during the month of December?

• December is the month in which tax is calculated,


so individuals have incentive to realise losses in
order to declare less profit and pay less tax
Forum Questions
1. What happens if the weighting function is non-linear? How
does it change the ways we value our prospects?
• The weighting function is non-linear, this property is called
subcertainty.
• Although, π(p) > p for low probabilities, there is evidence to
suggest that, for all 0<p<1, π(p) + π(1 - p) < 1.
• the sum of the weights associated with complementary
events is typically less than the weight associated with the
certain event
Forum Questions
2. I would like to clarify whether there are any cardinal differences between utility in EUT and
value in the Prospect Theory. While EUT deals with the expected value, Prospect Theory does
not deal with utility itself, and the Value function defines a model corresponding to the utility
function in the EUT, seemingly. Is this only for convenience, or am I missing something about
these terms?

If (𝑥,𝑝;𝑦,𝑞) is a regular prospect, then

Prospect Theory: 𝑉(𝑥,𝑝;𝑦,𝑞)= 𝜋(𝑝)𝑣(𝑥)+𝜋(𝑞)𝑣(𝑦)


Expected Utility Theory: 𝑈=𝑝𝑢(𝑥)+𝑞𝑢(𝑦)

You might also like