Scribd
Upload
3 views

Assignment 2

Uploaded by

liuzhe20030124
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

Assignment 2

Uploaded by

liuzhe20030124
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Time Series Analysis

ASSIGNMENT TWO

This assignment accounts for 10% of the total course mark. (Remarks for each problem is
indicated at the end of the question. This will be translated to the final course grade
proportionally.)

Some problems need to be completed by R. Please include your code at the end of the
assignment as Appendix.

Be sure to include your name and ID number.

Please hand in on time. Otherwise there will be a 50% discount on your remark.

1
1. Consider the following two models:
(1) 𝑍! = 𝛽" + 𝛽# 𝑡 + 𝑎! ,
(2) (1 − 𝐵)𝑍! = 𝑐 + (1 − 𝜃𝐵)𝑎! ,
assuming that 𝑎! ~𝑖. 𝑖. 𝑑. 𝑁(0,1), |𝜃| < 1 (for (2)).
(a) Given observations 𝑍# , 𝑍$ , … , 𝑍% , forecast 𝑍%&' , i.e., find 𝑍:% (𝑙) and
𝑉𝑎𝑟(𝑒% (𝑙)) for 𝑙 = 1,2, … , 𝐿. (10’)
(b) Show the similarity and differences in the nature of forecasts under the
two models. (Hint: Compare 𝑍:% (𝑙), 𝑍:%&# (𝑙), 𝑉𝑎𝑟(𝑒% (𝑙)) and
𝑉𝑎𝑟(𝑒%&# (𝑙)). What conclusions do you get?). (20’)

2. Let 𝑢! be an i.i.d. sequence with mean zero and variance 𝜎 $ , and let 𝑦! =
𝑢# + 𝑢$ + ⋯ + 𝑢( with 𝑦" = 0. Deduce that
𝑇 )#/$ F 𝑢( 1 1/2
, 0
D G → 𝑁 IJ K , 𝜎 $ L OP,
0 1/2 1/3
𝑇 )+/$ F 𝑦()#
where Σ indicates summation over t from 1 to T. Comparing this result with
Proposition 17.1, argue that
𝑊(1)
0 1 1/2
R U ~𝑁 IJ K , L OP,
T 𝑊(𝑟)𝑑𝑟 0 1/2 1/3
Where the integral sign denotes integration over r from 0 to 1. (Hint:
Exercise 17.1 in Hamilton(1994)) (20’)

3. Consider the data set: (1) R3.dat.


(a) Plot the series and comment on stationarity and general behaviour of the
series. (5’)
(b) Apply ADF test to see whether it has a unit root. (5’)
(c) Identify a time series model for the series. Briefly explain why. (5’)
(d) Estimate the model you specified from (c). (5’)
(e) Do diagnostic checking of the residuals. Briefly explain if the model is
adequate. (5’)

4. In E5_5.xlsx you will find time series data on maize_price, maize_yield,


wages, pig_price, pig_yield for China from 1867 to 1947.
(a) Are the series stationary? (10’)
(b) For nonstationary series, find out the integration order. (5’)
(c) Find appropriate model for each series. (5’)
(d) Obtain ten years ahead forecasts for each series and plot the figures. (5’)

You might also like