THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
15/16
STAT3301/4601/6002 Time Series Analysis
Assignment 1
(Due date: October 10, 2015)
Please answer at least Q1, Q2, Q3, two sub-questions of Q4, and three sub-questions of
Q5.
1

STAT3301/4601/6002 Time Series Analysis (2015-16)
Example Class 7
Model-building strategy
The Box-Jenkins method: specication (identication) estimation diagnostics
In this course, we focus on a large class of parametric models for both stationary
and no

STAT3301/4601/6002 Time Series Analysis (2015-16)
Example Class 8
Partial autocorrelation function (PACF) & sample PACF
Partial autocorrelation (PACF) (denoted by kk ): the correlation between Zt
and Ztk after removing the eect of the intervening variabl

STAT3301/4601/6002 Time Series Analysis (2015-16)
Example Class 10
The method of moments
Suppose that cfw_Zt follows an ARIMA(p, d, q) model, i.e. Wt =
follows an ARMA(p, q) model,
d
Zt = (1 B)d Zt
(1 1 B p B p )Wt = 0 + (1 1 B q B q )at ,
2
where cfw_a

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/4601/6002 Time Series Analysis
Chapter 4 Model specication
1
Model-building strategy
Box-Jenkins method for model-building strategy: Finding appropriate models for time s

Write your answers on this side only. Q. No. . .
§§ f.f[.,l:I:ff!‘ ifIﬁflﬂ'LII[
, , ,
.z
I
W r w
n
a
“v
M
3‘3»?
I
A
Q
‘3'
1—7
\J
WW)”me M, m M
5 (Ya .
W = kw;
g;
E;
m w M m
. A,
, W ;
= E (611;) E ( 4’34:th go 4* ‘ «was :2)
m iaéegweé

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/4601/6002 Time Series Analysis (2015-16)
Midterm Test
October 9, 2015
Time: 3:30 p.m. - 4:30 p.m.
This is an open-book test. You can use any references, but have to keep q

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/4601/6002 Time Series Analysis
Chapter 2 Stationary linear time series models
This chapter introduces the theoretical properties of time series determined by models, and t

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/4601/6002 Time Series Analysis
Chapter 1 Some fundamental concepts
What is time series
Time series data
Denition: A time series data is a sequence of observations ordere

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/6002 Time Series Analysis
Assignment 3
(Due date: November 30, 2015)
15/16
Please answer at least Q1-7 at least, and Q8-10 are elective.
1. Consider an AR(2) model
Zt = 1

THE UNIVERSITY OF HONG KONG
DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
STAT3301/6002 Time Series Analysis
Assignment 2
(Due date: November 14, 2015)
15/16
Please answer at least Q1-6 at least, and Q7-9 are elective.
1. Suppose we have 10 observations

STAT3301/4601/6002 Time Series Analysis (2015-16)
Example Class 6
ARIMA models
Dierencing operation: Zt = Zt Zt1 , where
= (1 B) is called the
dierence operator, and the new time series cfw_ Zt is called the rst dierence of
cfw_Zt or the dierenced sequ