FINM7403
Project FAQ
Semester 2, 2023
1
This document provides answers to some common questions about the written assignment.
Part A
What data is needed for Part A and where should we get it from?
You’ll need to source Australian macroeconomic data, which you can get from a variety of sources. The two
most common sources for Australian macro data will be the Australian Bureau of Statistics (ABS) and the
Reserve Bank of Australia (RBA).
You’ll need sources for your assessment of market expectations, which should come from news articles
and/or analyst reports that you can find with an online search.
You’ll also need to download historical index values for the industry indexes that you choose to put into your
portfolio from Capital IQ. Instructions for how to download these index values are available as a video tutorial
posted in the assignment folder of the Learn.UQ site.
You’re welcome to provide your own source for the risk-free rate, though the task description document
gives you a value that you can use.
What is meant by including a “potential economic shock” and its consequences?
This idea was described in Lecture 3. A potential economic shock just refers to a qualitative scenario that
would impact the economy in some way and that you could describe the consequences of. It may be a
statement about the overall movement of the business cycle, about political shifts, about resources shifts,
about movements in one specific economic indicator, or innumerable other options.
One important consideration in sector rotation is which businesses perform better (in a net profits sense)
under different stages of the business cycle. However, it’s not a profitable strategy to invest in businesses
that should have high profits in the expected next stage of the business cycle, because the prices of those
firms have already adjusted for market expectations. Instead, what matters is the difference between your
expectations and the market’s expectations. If the market believes that the economy is rising quickly, and
you believe that it is rising as well but not as quickly, then you should invest in firms that do better when the
economy isn’t rising, since those are the firms that will perform better than the market expected, while pro-
cyclical firms will perform worse than the market expected.
Thinking through a potential economic shock is a means of identifying a possible scenario that deviates from
market expectations. If you think through the consequences of a shock thoroughly enough, then you might
stand a chance to understand the economy in that situation better than the market does, and so can more
accurately determine how well various industries would perform. Your portfolio then represents a bet that
your proposed shock will occur, and the goal is to get a better price on this bet than is fair because you
understand the result of the bet better than the market.
Should our potential shock be the thing that we think is most likely to occur in the future?
No, it doesn’t have to be. A low-probability shock just represents a higher-risk bet. If you understand the bet
well enough by exploring the consequences deeply enough, you could get a better-than-fair price on this bet,
and it would be a high-risk high-reward play. You could reduce the risk of your portfolio by investigating
multiple potential shocks and hedging your portfolio by weighting it across industries that would be expected
to outperform in the event that one or other of the investigated shocks was realised.
FINM7403
Project FAQ
Semester 2, 2023
2
Is there a correct answer that we’re supposed to figure out for the economic analysis?
There is largely a correct answer for the current state of the economy, and it should not be too difficult to
figure out what it is. This is a simple enough, yet fundamental enough, part of the assignment that it’s
perfectly acceptable for you to seek out the opinions of other groups or students in the class without fear that
you’d be committing academic misconduct, though it’s not expected that course staff will solve this part of
the task for you.
There’s a little more leeway in your description of market expectations, though there are certainly some
incorrect answers that would be too difficult to justify.
There is no correct answer when it comes to a proposed future shock. The only criteria that you need to
meet is that your shock is:
i. Basically plausible, that is, that it has a high enough chance of occurring that a reasonable reader
would care about the outcome of it happening. There is no cut-off for probability here, but as a
negative example, the potential shock of an asteroid wiping out the east-coast of Australia would not
be considered a plausible enough shock to warrant an investigation of the economic consequences.
ii. Well-researched, that is, you’ve demonstrated that you did research in determining the likelihood of
the shock occurring and what its potential consequences would be.
iii. Explored in detail, that is, that your narrative for the occurrence of and consequences of your
potential shock includes all necessary detail for the reader to understand the likely state of the
economy if your shock were to occur.
What assets are we choosing from to add to the portfolio in Part A? Is it the stocks in high-
performing industries or the industries themselves?
In Part A you aren’t making any stock selection decisions. Your portfolio decisions involve only (1) selecting
which industry indexes to exclude entirely from your portfolio and (2) deciding how to weight each of the
remaining indexes in your portfolio. After selecting which industries to exclude, you should have 4-6 industry
indexes left in your portfolio.
What do we need to do with the index data?
After selecting which industries to include in your portfolio, you will need to do a Markowitz optimisation in
order to select the optimal weights. The inputs to a Markowitz optimisation are: (1) expected returns of each
asset; (2) variances of each asset; and (3) covariances between each asset. In this case the indexes are the
assets being weighted, so you need to calculate these statistics for each index in your portfolio.
Do we need to calculate statistics for every industry in order to make the decision of which
industries to exclude?
No, the decision to exclude industries is based on your macroeconomic analysis, not an analysis of the
historical returns and risks of those industries. Without your macro analysis to rely on to make adjustments to
calculated statistics, the historical data would not be helpful in making a decision on the basis of a sector
rotation strategy, so it’s not important to calculate it for every industry before excluding some industries. After
you select the 4-6 industries in your portfolio, you will then need to calculate expected returns, variances and
covariances of those industry indexes only.
FINM7403
Project FAQ
Semester 2, 2023
3
How do we actually calculate the statistics we need for each index?
You’re expected to estimate these statistics directly for each industry index, rather than by assuming that the
returns follow an index model. Although there are more sophisticated methods that would improve the
accuracy of your estimates, this task is intended as an exercise in the basic procedure of a Markowitz
optimisation, and so you are expected to simply calculate historical average returns and historical variances
and covariances. You should choose a historical time horizon to calculate averages over and should justify
your choice, but you don’t need to justify the choice to calculate averages of historical data to begin with.
Wouldn’t calculating statistics from historical data for each index average out the calculation across
different parts of the business cycle so that they would be poor estimates for the expected statistics
in a specific upcoming part of the business cycle?
Yes, and this should be acknowledged in your report. You aren’t required to make any sophisticated cyclical
adjustments to your averaging procedure. It’s sufficient to adjust the averages after calculating them based
on your macroeconomic analysis. For example, using a fictional industry called “Industry X”, if you calculate
the average historical return for the Industry X index as 6%, and you expect that Industry X will perform
better in the upcoming economy than the market expects it to, then you should adjust that value upwards,
perhaps to 8%. You should broadly justify the size of the adjustment, but your justification does not need to
be very specific. It’s enough to state whether you are making a small, mid-sized or large upwards or
downwards adjustment and to justify why that is. This is a simplification of the actual procedure of portfolio
optimisation to provide you with an opportunity to practice the basic processes.
How are we expected to choose the weights of our selected indexes?
You must undertake a Markowitz optimisation to determine the optimal weights. The Excel demo file for
Lecture 2 available on the Learn.UQ site gives an example for both the calculations of historical statistics
and the completion of a Markowitz optimisation using those calculated statistics. The live demonstration
where this file was initially created can be viewed at the start of the recording for Lecture 3.
Can we have negative weights in our portfolio?
There isn’t any rule in the task description that forbids short-sales, but you should be clear about how that fits
into your investment strategy and aligns with your qualitative assessments. A negative weight on an asset is
a bet that short-selling the asset will be profitable for the portfolio. If you have some sectors with negative
weights, it should be because you assessed those sectors to be poor performers and included them in your
portfolio for the express purpose of shorting them. You shouldn’t give negative weights to any assets that
you assessed to be high performers, or positive weights to any assets that you assessed to be poor
performers.
Is it okay that the optimisation process has assigned weights of zero to some of our sectors?
A weight of zero means that that asset is not included in the portfolio, so if you’ve stated in your report that
you believe this selection of indices should be included, then your qualitative assessment will contradict your
quantitative analysis. The first thing to check is to make sure that you’ve remembered to adjust the expected
returns and variances that you calculated using historical data. If you haven’t adjusted these values in line
with your future expectations, then you aren’t selecting weights that would be optimal in your predicted future
state. If you have remembered to adjust the statistics, then a weight of zero assigned by the optimiser means
you were inaccurate with either your (i) qualitative assessment of the sector; or (ii) adjustment amount. For
example, if you’ve determined that X sector should be a good buy because of your expectations of the
FINM7403
Project FAQ
Semester 2, 2023
4
future, and you’ve adjusted the expected return up by 2% because this value is roughly in line with your
expectations, and then the optimiser assigns a weight of zero, then either 2% is too low of an upward
adjustment to be in line with your assessment and you should increase it, or else your assessment that the
sector is a good buy is incorrect and the fact that you expect its future returns to be 2% higher than the
historical average is not enough of an increase to justify buying it. After running the optimiser and seeing that
a weight is set to zero (or, if you’ve allowed negative weights, that a sector you recommended to buy is
negatively weighted or a sector you recommended to sell is positively weighted), you must either go back
and adjust your qualitative analysis in light of the new quantitative information you received, or go back and
make different (still reasonable) adjustments to the statistics and try optimising again. You don’t have to
include in your report a description of your trials of this analysis, only the end result.
FINM7403
Project FAQ
Semester 2, 2023
5
Part B
What data is needed for Part B and where should we get it from?
You’ll need financial statements for each of the firms that you’re analysing. These can be obtained by doing
an online search for the annual reports of those firms.
You’ll also need betas for each of the firms, which are used in calculating the cost of equity capital for use in
a DCF analysis. Capital IQ includes market beta information for each stock. This beta is calculated from 5
years of market data, but it’s acceptable for the sake of simplification to use this beta without adjustment
even if you calculate the expected return of the market using a different time window. The beta value can be
found by searching for each stock individually on Capital IQ and looking for the value labelled ‘Beta 5Y’ on
the company profile page, which is the landing page that you’re shown when searching for a stock.
Finally, you’ll need historical data on GDP growth to calculate the average growth of the economy, and this
can be sourced from the ABS.
How do we know which companies we can choose from?
The firms that you select to increase or decrease in weight must come from any of the industries that you
selected in Part A. The only reason for this is that it makes no sense to vary the weight of a stock that isn’t in
your portfolio to begin with. The easiest way to find a selection of firms that you can choose from is to check
the index constituents of each of the ASX sector indexes for the industries in your portfolio on Capital IQ.
You can reach this page by first searching for the index, then selecting “Constituents” from the left-hand-side
navigation panel. The number of firms in each index varies significantly, but you can select any of these as
the firms that you’re analysing.
How much information is needed in our report about the analysis for each stock?
You need to include enough detail in your report about your final analysis that the reader understands fully
how you came to your conclusions. You do not need to include any description of steps of your analysis that
you later changed such that they no longer relate to your final conclusions. You can assume that the reader
has also taken FINM7403, so you don’t need to describe, for example, why or how you would do a DCF
valuation, only what the particular estimates and assumptions were that you made and why you made them.
You should provide screenshots of the final state of any numerical analysis for each method of valuation so
that it can be verified that you actually undertook the analysis as opposed to inventing the results. While you
must verify that you actually did perform the analysis, the focus of your writing should be only the unique
parts of your particular analysis, which are the decisions you made about inputs and the conclusions you
derived from the analysis.
What should we do if the different methods produce different valuations?
This is expected to occur, and the criteria for this part of the report requires you to acknowledge and justify
any differences. Remember that all valuation methods rely on different assumptions. It might be useful to
consider what those different assumptions are and how the differences in assumptions may have contributed
to different valuations. You should ultimately collate all of the output of your analyses into a qualitative
assessment of the strength of the stock as well as a price target range.
FINM7403
Project FAQ
Semester 2, 2023
6
General Questions
Do we need two introductions and two conclusions?
Yes, each of Part A and Part B is itself an independent report and requires an introduction and conclusion.
The only sections of your submission that will relate to both parts together will be the cover sheet and the
executive summary. You do not need to summarise the conclusions of Parts A and B together at the end of
your report – this is the purpose of the executive summary.
What should be included in the submission?
You should put your entire written submission into a single pdf file. A team representative should submit only
this single pdf file to both the Turnitin submission link and the Blackboard submission link. You do not need
to submit any Excel or other software or code files. Your report should document your use of external
software well enough for the reader to understand how you performed your analysis without having to verify
it for themselves using the raw software files.