THE UNIVERSITY OF NEW SOUTH WALES
SCHOOL OF ECONOMICS
TERM 1 2023
ECON2209 BUSINESS FORECASTING
FINAL EXAMINATION
1. Time Allowed: 24 Hours.
2. This is a Take-Home Exam. Your responses must be your own original work.
You must attempt this exam by yourself, without any help from others. By
submitting this exam you are declaring that you have NOT worked, collaborated,
or colluded with any other persons in the formulation of your responses. The
work that you are submitting is your OWN work.
3. Release date/time (via Moodle):
Saturday, 29 April 2023, 8:00am (Australian Eastern Standard Time)
4. Submission date/time (Via Turnitin):
Sunday, 30 April 2023, 8:00am (Australian Eastern Standard Time)
5. Submission details:
• A submission link is on the Moodle site under Final Exam.
• Submit your answer document in PDF format. Your file name should
follow this naming convention:
FE_your first name_zID_your last name_ECON2209.pdf
For example: FE_John_z1234567_Smith_ECON2209.pdf
• You get one opportunity to submit your file. Make sure that you submit
the file that you intend to submit.
• Your submitted answers should include the R code that you used.
6. Failure to upload the exam by the submission time will result in an immediate
late penalty of 5% from 8:01am on 30 April, followed by additional penalties of
5% per day or part thereof from the due date and time.
7. This Examination Paper has 4 pages: 2 cover pages and 2 pages with questions.
8. Answer all three questions.
PAGE 2 OF 2
9. Total marks available: 55 marks. This examination is worth 55% of the total
marks for the course.
10. Questions are not of equal value. Marks available for question sub-parts are
shown on this exam paper.
11. This Take-Home Exam paper cannot be copied, forwarded or shared.
12. Students are reminded of UNSW’s rules regarding Academic Integrity and
Plagiarism. Plagiarism is a serious breach of ethics at UNSW and is not taken
lightly. For details see Examples of plagiarism.
13. This Take-Home Exam is an open book/open web exam. Further information is
available by clicking here.
• You are permitted to refer to your course notes, any materials provided by
the lecturer, books, journal articles, or tutorial materials.
• You are required to cite your sources and attribute direct quotes
appropriately when using external sources (other than your course
materials).
• It is sufficient to use in-text citations that include the following information:
the name of the author or authors; the year of publication; the page number
(where the information/idea can be located on a particular page when
directly quoted), For example, (McConville, 2011, p.188).
• When citing Internet sources, please use the following format: website/page
title and date.
• If you provide in-text citations, you MUST provide a Reference List.
14. Use of AI tools such as ChatGPT are prohibited. In cases where use is
detected, penalties will apply.
15. Students are advised to read the exam paper thoroughly before commencing.
16. The Lecturer will be available online (via Moodle) after the exam paper is
released for a period of two hours.
UNSW ECON2209 Final Exam
2023
Remember to type ‘library(fpp3)‘ in the R Studio Console before you start.
Answer all three questions
Question 1 [20 marks]
This question uses data from the Australian Bureau of Statistics (ABS) on consumer prices, from ABS
Catalogue 6401.0, Consumer Price Index, Table 9. CPI: Group, Sub-group and Expenditure Class, Index
Numbers by Capital City: In particular, this question will use the inflation index for “Furnishings, household
equipment and services; Sydney”, series ID A2325986F. Access the series using the following commands:
library(readabs)
myseries <- read_abs_series("A2325986F") %>%
mutate (Quarter = yearquarter (date)) %>%
as_tsibble (
index = Quarter,
key = c (series_id)
) %>%
filter(year(Quarter)>=2002)
a. [10 marks] For this data series, create a training dataset (myseries_train) consisting of
observations before 2017. Check that your data have been split appropriately by producing a
plot of myseries_train and myseries in one figure. Then calculate seasonal naïve forecasts
using SNAIVE() applied to the training data and check the residuals. Produce and plot
forecasts for the period of the test data, and plot these on the same figure as your full data
set. Include the forecast prediction intervals. Using RMSE and MAPE, compare the model’s
performance against the training data set relative to its performance against the test data
set. Be sure to discuss all your results.
b. [10 marks] Now fit a time series regression model to the full data series, using a linear trend
and seasonal dummies as predictor variables. Plot the fitted model along with the actual
data. Create a dummy variable to model the sharp fall in the series in quarter 2 of 2020 and
add this to your regression model, then plot the fitted model along with the actual data.
Now add also a knot in the time trend at quarter 2 of 2020, then plot the fitted model along
with the actual data. For each of these three models, using the plots, discuss the performance
of your model relative to the actual data. Report the AIC, AICc and BIC for each of the
models. Which model is preferred and why? Forecast two years ahead using your preferred
model and produce a plot of the actual data and the forecasts, including prediction intervals.
Are the model’s forecasts reasonable? Discuss.
1
Question 2 [20 marks]
Use your (non-seasonally adjusted) expenditure volume data from the Course Project:
Plot your series with the series name in the title. Develop an appropriate dynamic regression
model with Fourier terms for the seasonality. Use the AICc to select the number of Fourier terms
to include in the model. (Use any appropriate Box-Cox transformation you identified in the
Course Project, or should have identified in the Course Project.) Describe your model, and check
if the residual series is white noise. Forecast five years ahead using your preferred model and
produce a plot of the actual data from 2010 and the forecasts, including prediction intervals. Are
the model’s forecasts reasonable? Discuss.
Hint: For each K, try including e.g. pdq(0:2, 0, 0:2) + PDQ(0:1, 0, 0:1) in your model specification to
help with quickly considering a lot of different models.
10 marks for use of appropriate methods, 10 marks for a sensible, clear description and
discussion.
Question 3 [15 marks]
Use the Australian real GDP data you used in Problem Set 2 (series ID A2302459A):
Use ARIMA() to find an appropriate ARIMA model for this series. What model is selected? Write
the model in terms of the backshift operator. Check that the residuals look like white noise. Plot
forecasts with prediction intervals for the next two years, along with the actual data from 2015.
Now produce the same syle of plots for forecasts from an ARIMA(0,1,0) model with drift and
forecasts from an ARIMA(2,1,2) model with drift. Describe these models, and compare their
forecasts with the forecasts of the automatically selected model. Discuss the differences and
explain which model you would select.
5 marks for use of appropriate methods, 10 marks for sensible, clear description and discussion.