SEEM 4480 Decision Methodology and Applicatons
Xuedong He, Fall 2017
Suggested Solution to Homework 4
1. Denote the arrival time of the first bus as x minutes. Then, x follows exponential
distribution with parameter 1/20. Denote the arrival time of Mary a

SEEM4480DecisionMethodology
andApplications
Tutorial7
Outline
Describe the decision environments of certainty and
uncertainty
Construct a payoff table and an opportunity-loss table
Define and apply the expected value criterion for decision
making
Comp

Tutorial 1
YUAN, Quan
Background: expected value
A fruit seller example
Undamaged mango
Damaged mango
Profit
$4
$1
Probability
80%
20%
What is the expected profit for a stock of 100 mangoes ?
0.8 x 100 ($4) + 0.2 x 100 x ($1) = 320 + 20 = $340
random vari

SEEM 4480
Decision Methodology and
Applications
Tutorial 3
YUAN, Quan
1
Introduction
EMV=$14.5
Game 1
(0.5)
(0.5)
EMV=$50 (0.5)
Game 2
(0.5)
Payoff
$30
-$1
$2,000
Which game would you
choose, game 1 or
game 2?
-$1,900
If EMV is the basis for the decision,

SEEM4480 Decision Methodology
and Applications
-Bayesian Analysis of the
Normal Distribution
Tutorial 10
Benchmark: The maximum likelihood estimate
for the normal distribution known variance
Let yi ~iid N( , 1) for i = 1, , n. The likelihood of is:
1
L( |

SEG4480 DECISION
METHODOLOGY AND
APPLICATIONS
Tutorial 2
Quan YUAN
AN EXAMPLE
| Suppose
you are considering whether or not to open
a Starbucks on CU campus. To open it, it will cost
150,000 HKD.
| There two possible results, one is that this new cafe
can

SEEM 4480
Decision Methodology and
Applications
- Value of Information
Quan Yuan
Information Collection - Key
Strategy
Motivation
To reduce uncertainty which makes us choose
second best solutions as insurance
Concept
Insert an information-gathering stag

SEEM 4480
Decision Methodology and
Applications
Tutorial 4
Yuan Quan
Outline
Indifference Curve
Insurance Market
Random Variables
Sophies Choice
Sophie eats chocolate bars and drinks
soda.
She can buy 10 chocolate bars or 20
sodas or some combination

Tutorial 4 for SEEM 4480
Tutorial 4 for SEEM 4480
Tutor: Ms. Lu, Bojun
February 18, 2013
Tutorial 4 for SEEM 4480
Content Review
There are at least two reasons for St. Petersburg paradox:
The first is that the mathematical term expectation should not
be i

Tutorial 2 for SEEM 4480
Tutorial 2 for SEEM 4480
Tutor: Ms. Lu, Bojun
January 28, 2013
Tutorial 2 for SEEM 4480
Content Review
EMV - The Expected Monetary Value Strategy:
L (d ) =
X
L (d , )p ().
The naive bayes model, Bayes learning and log odds ratio s

Tutorial 5 for SEEM 4480
Tutorial 5 for SEEM 4480
Tutor: Ms. Lu, Bojun
February 25, 2013
Tutorial 5 for SEEM 4480
Content Review
(Weak substitution) If P1 , P2 and P are any three reward
distributions, then for all probabilities , 0 < < 1,
P1 P2 P1 + (1 )

Tutorial 1 for SEEM 4480
Tutorial 1 for SEEM 4480
Tutor: Ms. Lu, Bojun
January 21, 2013
Tutorial 1 for SEEM 4480
Content Review
In probability theory and statistics, Bayes theorem
(alternatively Bayes law) is a theorem with two distinct
interpretations.
I

Tutorial 3 for SEEM 4480
Tutorial 3 for SEEM 4480
Tutor: Ms. Lu, Bojun
February 4, 2013
Tutorial 3 for SEEM 4480
Content Review
Rollback Method:
The rollback method starts at the terminal nodes of the tree
and works backward to the initial decision node,

Tutorial 3 for SEEM 4480
Tutorial 3 for SEEM 4480
Tutor: Ms. Lu, Bojun
February 4, 2013
Tutorial 3 for SEEM 4480
Content Review
Rollback Method:
The rollback method starts at the terminal nodes of the tree
and works backward to the initial decision node,

Tutorial 6 for SEEM 4480
Tutorial 6 for SEEM 4480
Tutor: Ms. Lu, Bojun
March 4, 2013
Tutorial 6 for SEEM 4480
Content Review
Expected Value of Perfect Information (EVPI): Maximum EMV
with perfect information - Maximum EMV with current
information.
The exp

SEEM4480
Tutorial 7,8,9
2013 March
SEEM4480
2
1 Paired random variables - discrete case
X; Y; Z are three random variables that could take the value of 0 or 1. In the following, (x; y; z) means
(X = x; Y = y; Z = z).
The possible events are f(0; 0; 0)g,f(

Decision Methodology and
Application
Tutorial 5
Yuan Quan
Example: Parameter Estimation
There is a coin, how do you estimate the
probability (p) that it turns out to be a
head after tossing.
Example: Parameter Estimation
Suppose you only toss once, and

Integer Programming
Part 2
Solving Integer Programming Problems
Even though there is a finite number of feasible
solutions for IPs, it is in general computationally
intractable to use complete enumeration to solve such
problems
For a problem similar to

SEEM 4480 Decision Methodology and Applicatons
Xuedong He, Fall 2017
Homework Set 6
(Due Apr. 26)
Assignment
1. The proportion p of individuals in the population that have attribute X is
unknown. Suppose that on the basis of a pilot sample p is given a be

SEEM 4480 Decision Methodology and Applicatons
Xuedong He, Fall 2017
Homework Set 5
(Due Apr. 13)
Assignment
1. (a) Consider a two-act problem with linear value where the corresponding
utility functions are given by
u(a1 , s) = 840s 10,
u(a2 , s) = 360s +

SEEM4480: Decision Methodology and Applications
Lecture 2: An Informal Treatment of Foundations
Instructor: Prof. Xuedong HE
Department of Systems Engineering & Engineering Management
The Chinese University of Hong Kong
Spring 2017
Notes by Prof. Xiting G

SEEM4480
Decision Methodology and Applications
Lecture 1: Introduction
Instructor: Prof. Xuedong HE
Department of Systems Engineering & Engineering Management
The Chinese University of Hong Kong
Spring 2017
Notes by Prof. Xiting Gong in Fall 2015 (CUHK)
S

SEEM4480 - DECISION METHODOLOGY & APPLICATIONS
Fall 2015
Solution of Assignment 1
September 30, 2015
1. Let c and c denote consequences Win and Lose respectively. Let the symbol hi denote
a basic reference lottery that gives c with canonical chance and c

THE CHINESE UNIVERSITY OF HONG KONG
Department of Systems Engineering & Engineering Management
2016/2017
2nd Term
Course Code & Title: SEEM4480: Decision Methodology and Applications
Instructors and Teaching Assistants
Instructor
Name
Office
Telephone
Ema

SEEM4480: Decision Methodology and Applications
Lecture 10: Preposterior Analysis: Value of Sample
Information
Instructor: Prof. Xuedong HE
Department of Systems Engineering & Engineering Management
The Chinese University of Hong Kong
Spring 2017
Notes by

SEEM4480: Decision Methodology and Applications
Lecture 9: Bayes Rule
Instructor: Prof. Xuedong HE
Department of Systems Engineering & Engineering Management
The Chinese University of Hong Kong
Spring 2017
Notes by Prof. Xiting Gong in Fall 2015 (CUHK)
SE

Welcome, Students!
Elementary Cantonese for Chinese
Speakers
CLCC1703
Teacher: Mira Wong
Yale-China
Chinese Language Center (CLC)
CLCC1703-C
Elementary Cantonese for Chinese Speakers (Lecture)
3 credits
Wed 14:30 - 16:15 UCC_108
(T.C. Cheng Bldg 108)
http

JIANG YUFEI (A0130428N)
NAME: JIANG YUFEI
MATRIC NO: A0130428N
CLASS: E15
TUTOR: Happy Goh
ES1531/GEK1549/GET1021 Critical Thinking and Writing
ASSIGNMENT 1
NATIONAL UNIVERSITY OF SINGAPORE
2015/2016 Semester 2
I understand what plagiarism is and have ens

HEURISTICS
Tutorial 10
11 November 2016
IE4210 Operations Research II
Problem 1
Four jobs must be processed on a single machine. The
number of days needed to complete each job, the earliest
a job can be completed, and the due date for each job are
given

Eigenvalues and Eigenvectors
If Av=v with v nonzero, then is called an eigenvalue of A and v is called an
eigenvector of A corresponding to eigenvalue .
Agenda: Understand the action of A by seeing how it acts on eigenvectors.
(A-I)v=0 system to be satisf

THE CHINESE UNIVERSITY OF HONG KONG
Department of Systems Engineering & Engineering Management
2016/2017
2nd Term
Course Code & Title: SEEM4480: Decision Methodology and Applications
Instructors and Teaching Assistants
Instructor
Name
Office
Telephone
Ema

IE4210 Operations Research II
AY2016/2017 Semester 1
Tutorial 2
For discussion in tutorial on 02/09/16:
1.
2.
3.
Problem Set 9.2A Q5, pg 503.
Review Problem Set 9A Q2, pg 552-553. For this problem, consider break
points 0, 0.25, 0.5, 0.75 and 1 when appro

1. Problem description
Long queuing time for checkout in the supermarket during peak hours is a common
phenomenon in Singapore market. People usually wait for about 10 minutes in the queue to be
served. To solve the congestion problem, supermarkets instal

Tutorial 6 for SEEM 4480
Tutorial 6 for SEEM 4480
Tutor: Ms. Lu, Bojun
March 4, 2013
Tutorial 6 for SEEM 4480
Content Review
Expected Value of Perfect Information (EVPI): Maximum EMV
with perfect information - Maximum EMV with current
information.
The exp

Tutorial 5 for SEEM 4480
Tutorial 5 for SEEM 4480
Tutor: Ms. Lu, Bojun
February 25, 2013
Tutorial 5 for SEEM 4480
Content Review
(Weak substitution) If P1 , P2 and P are any three reward
distributions, then for all probabilities , 0 < < 1,
P1 P2 P1 + (1 )