ECMT2130 Financial Econometrics Semester 2, 2023
GROUP PROJECT
DUE: 11.59pm Friday, 3rd of November 2023
Academic Honesty
Academic honesty is a core value of the University, and all students are required to act
honestly, ethically and with integrity. The consequences of engaging in plagiarism and
academic dishonesty, along with the process by which they are determined and applied, are
set out in the Academic Honesty in Coursework Policy 2015. Under the same policy, as the
lecturer, I must report any suspected plagiarism or academic dishonesty.
Instructions
• This is a group assignment which accounts for 20% of your final mark.
• You can either hand-write or type your answers, but please compile all your answers in
one single PDF file and submit it via a file upload in Canvas. You can only submit your
work once, so please double check everything before submitting.
• There are 15 questions in this assignment, and please attempt all questions.
• I will randomly select 5 questions to mark, and each question is worth 4 points. The total
number of points of this assignment is therefore 20. The grading will be based on the
completion and general quality of your submission.
• Refer to the group spreadsheet (link here: Spreadsheet) to find out which ASX200
company your group has been allocated to. Failure to work with the correct series will
result in a 50% loss of the entire mark of the assessment, i.e., your maximum mark will
be 10 (instead of 20).
• Answer all questions in a neat PDF document (no other extensions accepted). Use Times
New Roman font size 12 throughout the report and normal margins. Make sure any
pictures included in the document are pasted correctly: your report should look neat, clean
and easy to navigate.
• Only the group leader is required to submit the final report. Please, make sure to include a
cover sheet with the SIDs of the group members, their signatures and the series assigned
to the group.
• Based on the University late policy, a late submission is subject to a penalty of 5% (of the
total points) per calendar day; and work submitted more than 10 days after the due date
will receive a mark of zero.
• You are free to use any econometric software and/or Excel. It is also fine to work with a
combination of them. If you are using R, submit your script. If you are working with
Excel, make sure to include the spreadsheet in xslx format with brief explanations of
which question you are doing. If you are working with anything else, make sure to
include commands and results in a neat document. Your software work file should be
separate from the main report with the solutions (so, you’ll likely submit two files).


2

QUESTIONS

Throughout the assignment, I will symbolise the series of (close) share price as and the
corresponding series of returns with . The time span you should use is the one specified in
the “Data.R” script provided, i.e., from the 1st of January 2020 until the 1st of October 2023.
Using your assigned financial time series, answer the following questions:

1) Provide a brief overview of what the company you have been selected does. Make sure to
include the goods/services it commercialises, number of clients, where it operates in the
world, its main competitors and its last financial figures. Do not surpass more than half a
page on this question.
2) If the Efficient Market Hypothesis (EMH) is true, do you expect to find a good model to
forecast the conditional mean of ? Briefly explain.
3) Plot . Make sure to include labels on the y and x axes, as well as a title. Describe its
main features from a classical decomposition perspective.
4) Calculate the average, standard deviation, minimum, maximum, skewness and kurtosis of
your series of prices and return series over the time span specified. Place them all on a neat
table, preferably prepared in Excel or equivalent (do not copy and paste R output). Interpret
the kurtosis of your series.
5) Propose a histogram for and another for . Using the graphs and the basic statistics
obtained in question 4, indicate whether the sampling distribution of price and return are
visually likely to be normally distributed. Feel free to overlay a normal distribution on top of
the histograms to strengthen your argument and/or use statistical tests of normality.
6) Calculate the Sharpe ratio for and for the market proxy (ASX200 returns). You will
need to find the return of the risk-free asset for the relevant period. Briefly justify why you
selected such a risk-free rate.
7) Calculate the alpha and beta of your assigned share. Interpret both values. Do you have
evidence pro or against the CAPM in your case? Make sure to indicate the null hypothesis of
your tests in the explanation. Hint: be careful here with the number of observations for both
dependent and independent variables.
3
8) Fit the following models to :
(i) Drift.
(ii) Mean.
(iii) Naïve.
(iv) 5-MA.
Evaluate, using the in-sample RMSE, which model fits the data best. Plot a graph of the
series of prices and fitted values of your best model in the same picture. Forecast the next two
days of data using your favourite model.
9) Fit the following models to :
(i) Simple Exponential Smoothing (SES).
(ii) Holt’s linear trend.
Which model fits the data better using the MAPE? Justify your choice. Forecast the next two
days of data using your favourite model. Make sure to include 95% confidence intervals
around the forecasts.
10) Is a stationary series? Use the KPSS test, as well as the ACF plot to justify your claim.
Apply an appropriate level of differencing, if necessary. If you do, show that the transformed
series is now stationary using the KPSS test.
11) Using the ACF and PACF of the (potentially differenced) return series, propose a suitable
ARMA model. Explain how you obtained your answer and write out the model specification.
Compare this model with a simple model regressed on a constant only (with specification
= + ). Use an information criterion of your choice to decide.
12) Using your preferred ARIMA(p, d, q) model, produce forecasts for the next 5 days. Make
sure to calculate the 95% forecasting intervals.
13) Using actual data from Yahoo Finance, pick the model (from Q8, Q9 and Q12) that best
forecasted the next five business days of . Contrast this with what you expected in Q2.
14) Using , propose an ARCH(1), ARCH(2) and ARCH(3) model. Are the parameter
conditions met? Which one seems most appropriate for your data? Justify your answer. Plot
the fitted variance of the returns.
15) Using , propose a GARCH(1,1) model. Plot the fitted variance of the returns. Are the
period(s) of high observed volatility in the graphs in Q3 consistent with the predicted
volatility generated by the GARCH(1,1) model?