STAT 4006: Categorical Data Analysis
Assignment 6
Academic year 14/15, rst term, Chapter 6
Due date: 6:00pm Dec 3(Wednesday), 2014. Assignment Box at LSB 1/F
1. There are 9 dierent hierarchical loglinear models can be t to a contingency table
with three v
STAT 4006: Categorical Data Analysis
Assignment 4
Academic year 14/15, rst term, Chapter 4
Due date: 6:00pm Nov 7(Friday), 2014. Assignment Box at LSB 1/F
1. Access the horseshoe crab data. Let Y = 1 if a crab has a least one satellite, and let
Y = 0 othe
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 8
1. (Problem 3.3 from Agresti (2007) Refer to Example 2.13 of the
Chapter 2 lecture notes (Slide 79) and let x = mothers alcohol consumpti
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 8 - Solutions
1. (Problem 3.3 from Agresti (2007)
(a)
= 0.00255 + 0.00109(alcohol).
(b) Estimated probability of malformation increases f
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 6
1. (Question 3 from 2002-3 Final) Consider the following table of data:
Smoking Status
Never smoked
Former smoker
Current smoker
Breathin
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 1
1. (Problem 1.2 in Introduction to Categorical Data Analysis, 2nd Ed. by
Agresti) Which scale of measurement is most appropriate for the
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 4
1. Show that the test statistic for the test of independence (or homogeneity)
for 2 2 contingency tables is given
X2 =
n(n11 n22 n12 n21
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 5
1. (Problem 2.39 from Agresti (2007) True, or false?
(a) In 2 2 tables, statistical independence is equivalent to a population
odds ratio
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 3 - Solutions
1. Following the example of Slide 25 of the Chapter 1 notes, try to find the
MLE under these null hypotheses.
(a) H0 : 1 + 2
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 6 - Solutions
1. (Question 3 from 2002-3 Final)
(a) They are ordinal.
(b) There are 177381 concordant pairs and 91573 discordant pairs in t
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 1 - Solutions
1. (Problem 1.2 in Introduction to Categorical Data Analysis, 2nd Ed. by
Agresti) Which scale of measurement is most appropri
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 2 - Solutions
1. (Problems 1.12 and 1.13 in Introduction to Categorical Data Analysis,
2nd Ed. by Agresti)
(a) The estimated standard error
STA 4006: Categorical Data Analysis
Exercise 2
Academic year 07/08, first term, Chapter 2
1. According to recent UN figures, the annual gun homicide rate is 62.4 per one million
residents in the United States and 1.3 per million residents in the UK.
a. Co
STA 4006: Categorical Data Analysis
Exercise 3
Academic year 07/08, first term, Chapter 3
1. The data in the table are obtained from 322 couples with ages 32 to 35. Analyze this
set of data.
(Z=1)
Years of marriage < 5
X
Y
Blood pressure (Wife)
Abnormal
B
STA 4006: Categorical Data Analysis
Exercise 1
Academic year 07/08, first term, Chapter 1
1. Consider the statement, Please tell me whether or not you think it should be possible
for a pregnant woman to obtain a legal abortion if she is married and does n
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 7 - Solutions
1. (Problem 2.39e from Agresti (2007) False.
2. (Problem 2.37 from Agresti (2007)
(a) 0.18 for males, 0.32 for females.
(b) 0
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 2
1. (Problems 1.12 and 1.13 in Introduction to Categorical Data Analysis,
2nd Ed. by Agresti) To collect data in an introductory statistic
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 3
1. Following the example of Slides 24 and 25 of the One-way Tables notes,
try to find the MLEs of the multinomial parameters under these
STAT 4006: Categorical Data Analysis
Assignment 5
Academic year 14/15, rst term, Chapter 5
Due date: 6:00pm Nov 21(Friday), 2014. Assignment Box at LSB 1/F
1. In the rst nine decades of the twentieth century in baseballs National league, the
percentage of
Svff LZI 6%
() Like/{W fmvél'tm and MLE
Z) H7pow5¢ fob/1m}
a) wow w
>
b) Pemgzm : fed:
X2 / Z ( 5JL
E
a) LR/l
0
df 7 :tfvf (Le/Wf afWhr/k PM
S: if of (pa/r0» (Mob/r HOUHI " it of 170W (Jqu Ho
¢ 34 of M476 (7W under How, " it? of 0M9?
6-3. c4, we: mm}, 4
Chapter 4. Generalized Liner Models
4.1 Generalized Linear Model
4.2 Generalized Linear Models for Binary Data
4.3 Generalized Linear Models for Count Data
4.4 Extension of Notation in Generalized Linear Models
4.5 Generalized Linear Mixed Models*
1
4.1 G
:tgl,lipli1'i
:r:f~illEP
Convrie.ht Reservtd
he Chinese University of Hong Kong
Course EX8.11lination 2nd Tenn, 2012-13
STAT4006 Categorical Data Analysis
Course Code & Title:
M
2
+~
o
hours
Time allowed
minutes
l~.-5Ift
Seat No. :
Student LD. No.
Final
tttlJ ~-!Not to be taken away
~ -
Ji (#-
4-
Ji) Page 1 of tr
.x. A.
W 5% tf
~
The Chinese University of Hong Kong
.:.0
Course Examination 1st Term, 2013-14
*t EI ~ll.t~
Course Code & Title;
STAT 4006 Categorical Data Analysis
+*
*M
Time allowed
"Hi
o
h
~
~
~
t
3:.
*
- Jf (#- 3 Ji) Page J of 3
~
J:Jifi'/cfw_FfrH
~'1~ralr:1J
Cooyrie:ht Reserved
The Chinese University of Hong Kong
-=-0
-+.f.-.1f-~
L~ff ~
Course Examination 1st Term, 2010-11 .
#Me~.&.t~
STAT 4006
Categoric~l
Data Analysis
Course Code & Titl
Chapter 6. Loglinear Models for Contingency Tables
6.1 Introduction
6.2 Loglinear Model for Two-way Tables
6.3 Loglinear Models for Three-way Tables
6.4 The Loglinear-Logit Connection
6.5 Model Selection and Comparison
1
6.1 Introduction
In Chapter 4, we
THE CHINESE UNIVERSITY OF HONG KONG
Department of Statistics
STAT4006: Categorical Data Analysis
Exercises for Week 4 - Solutions
1. The proof is pure computation. We want to show
n(n11 n22 n12 n21 )2
(n11 n1+ n+1 /n)2
=
n1+ n2+ n+1 n+2
n1+ n+1 /n
(n12 n1