SU Risk and Prior Outcome and Managerial Decision-Making Presentation

Description

2 attachmentsSlide 1 of 2attachment_1attachment_1attachment_2attachment_2.slider-slide > img { width: 100%; display: block; }
.slider-slide > img:focus { margin: auto; }

Unformatted Attachment Preview

The decision maker
matters
Individual vs. Group Behaviour in
Experimental Beauty Contest Games
Martin G. Kocher and Matthias Sutter
Kocher, M. G. and Sutter, M. (2005), The Decision Maker Matters: Individual
Versus Group Behaviour in Experimental Beauty-Contest Games.
The Economic Journal, 115: 200–223. doi: 10.1111/j.1468-0297.2004.00966.x
The decision maker matters
In economic theory decision making agents usually modelled as an individual
BUT: Many important decisions are made by (small) groups

E.g. families/households, boards of directors, legislatures, committees, etc.
Consumption spending
Monetary policy
Company’s business strategy
Military decisions
Research questions
Looking at the influence of groups vs. individuals on the rationality of decision
making in an interactive context:
1.
Are groups more rational decision makers in the sense that their decision is
closer or converge faster to the equilibrium prediction?
2.
Do groups outperform individuals in terms of payoff when competing
against individuals?
FOCUS: Rationality & Learning
Individual vs. Group reasoning
Intellective Tasks – non-interactive

Typically groups perform better on purely
intellective tasks, where the “truth wins”
Groups are able to share knowledge
on average groups are closer to the
correct solution/ more often get to the
right solutions than individuals
Judgemental Task – interactive

No systematic difference in groups and
individual’s performance
Groups experience :

“Group think” – conformity & selfcensorship
“Risky-shift” – polarise individual
judgement
Reasoning – Hypothesis
Hypothesis #1:
Group guesses are closer to the game-theoretic equilibrium because they apply
deeper levels of reasoning – intellective
Hypothesis #2:
If groups compete directly against individuals, groups should win the game more
often
Individual vs. Group learning – Hypothesis
According to information load theory:

Groups are able to process high information load in intellectual task better
than individuals
Hypothesis #3:
Convergence towards the equilibrium is faster for groups than for individuals
Why Beauty-Contest Game


Classic Beauty contest game with p=?, repeated for 4 rounds
The beauty-contest game does not confound effects of rationality & learning
with the effects of social preferences like inequality aversion, fairness or
reciprocity
Loss or risk aversion cannot occur in the beauty contest game
The beauty-contest game is at the same time interactive and easy to explain
The beauty-contest game includes an intellective and judgemental task
The Experiment – Subject Pool
Experiment 1: Homogeneous decision makers

individual vs. individual, group vs. group
Classroom experiment
140 participants
Freshman at University of Innsbruck
Enrolled in Economics course
2 parallel sessions with 17 and 18
decision-makers in each treatment
Groups consistent of 3 subjects
Payout for winning number $10 per
subject
Experiment 2: Heterogeneous decision makers

Individuals vs. groups
Lab experiment
60 participants
Freshman at University of Innsbruck
Enrolled in Economics course
3 sessions with 4 observation units each
(2 individuals + 1 group)
Groups consistent of 3 subjects
Payout for winning number $5.50 per
subject
Experiment 1 – Design

Random assignment to experimental treatment and group
Subjects received written instructions, which were read out loud
Option to privately ask questions
After each round, chosen numbers were read out loud and the mean,
p*mean and the winning number announced
5 minutes decision time after each round
Subjects were asked to write down their reasoning for each decision
Sessions lasted up to 40 minutes
Private payout at the end of session
Cumulative Frequencies of Guesses
Experiment 1 – Results
Round 1

Mean and median far away from equilibrium
No significant difference in the mean of individuals and groups
According to the explanation sheets, reasoning of choice are not much
different for individuals and groups
Reject Hypothesis #1 for first round
Experiment 1 – Results
Round 2, 3 & 4

Chosen number decline significantly over time for individuals & groups
Groups converge faster to equilibrium than individuals
From round 1 to 2 behaviour difference significant at p < 0.01 level From round 2 to 4 behaviour difference significant at p < 0.001 level Groups understand intellectual aspect of game faster Supports Hypothesis 3 Experiment 1: Depth of Reasoning - Results - - Subjects do not choose exact equilibrium Groups reach reasoning levels 3 and 4, Individuals stay at level 2 For round 2 to 4 groups apply deeper level of reasoning, significant at p Purchase answer to see full attachment Explanation & Answer: 20 Slides Tags: RISK AND PRIOR OUTCOME AND MANAGERIAL DECISIONMAKING User generated content is uploaded by users for the purposes of learning and should be used following FENTYESSAYS.COM ESSAY's honor code & terms of service.