You are on page 1of 5

COMPUTER ASSIGNMENT

INSTRUCTIONS
FINS2624 Semester 2, 2015
Please read these instructions carefully before you start.

Groups
You are encouraged to work on this assignment in groups consisting of up to three students. If you
prefer, however, you can work alone.
You need to register your group in the Assignment section that has been created on Moodle.
Simply go to the Enrol in a Group link and follow the instructions. The window for registrations will
close on 22 September. Note, for ease of marking and results distribution those opting to work as
individuals are required to register as a group as well.

Submission
Your group should submit both a completed Excel Workbook Computer Model and a Group Report
(Word or pdf document). These are to be uploaded through the Assignment Submissions link
which is also located in the Assignment section of the course Moodle site.
Please name your files Assignment - Student ID, where Student ID is the Student ID (including the
letter) of the first student listed in your group. The deadline for the assignment to be received is 9
a.m. EST, 6 October 2015.
Make sure to enter the Student IDs (including the prefix letter) and names of all students in your
group in:
The appropriate cells (F2:F7) on the Excel Answers tab.
In the appropriate section of the Group Report template.
The answers sheet automatically summarises key figures that you are required to enter into the first
section of your group report.

The Excel sheet


You can download the Excel-sheet from the same place on Moodle where you found this document.
Each non-empty cell in the sheet has a colour which depends on its contents. Green cells contain
given data. Blue cells contain given functions. You should not change the values of either green or
blue cells. Orange cells are empty, and the assignment consists in filling these out correctly. When
you download the workbook some sheets are in protected mode, so that you will only be able to

Assignment Instructions

Page 1 of 5

change the cells you are supposed to change. You can disable the protection by right-clicking the
name of the sheet at the bottom of the workbook and select Unprotect Sheet.
There are any number of ways to come up with the correct solution, including making additional
calculations in other cells or sheets or using some completely different software. You are, however,
required to solve this assignment using solely the provided Excel file and by manipulating the orange
cells alone. Please do not enter data anywhere but in the orange cells.
The Returns 2Jan09-2Jan15sheet contains daily return data for six ETFs and a risk-free asset for
trading days in the period January 2, 2009 to January 2, 2014. The corresponding prices are on the
Prices 31Dec08-2Jan15 sheet. These securities are taken to represent the entire market. The risky
returns are actual realised adjusted daily returns, but the risk-free return has been set to a constant
to reflect the setup we had in class.

The Excel Solver


At several points in the assignment you will have to solve optimisation problems. If you know how to
do this in Excel you may skip this section.
The tool used to do solve optimisation problems in Excel is an Add-in called the Solver. You may not
have this activated in Excel, in which case you will have to activate it. In Excel 2010 this is done by
clicking File - Options - Add-Ins. Select Excel Add-ins from the drop down menu at Manage and click
<Go>. Make sure that Solver Add-in is checked and click <OK>. The solver will now be accessible
under the Data-tab.
The process is similar in other versions of Excel. Instructions to load the solver in Excel 2013 can be
found here. Instructions to load the solver in Excel 2007 can be found here.
The solver allows you (among other things) to solve maximization problems numerically. To do this
you select a cell whose value you want to maximize. This cell should contain a function (your
objective function) that depends on the values in some other cells. You can then specify which of
these cells (your arguments) the Solver is allowed to change in order to find the optimum.

Example
To see how this works go to the Solver example-sheet in the workbook. Cell C4 contains a function
that depends on the values in cells C5 and C6. To make excel find the values in C5 and C6 that
maximizes the value in C4, select cell C4 and click the Solver button. At Set Objective you can enter
the cell containing your objective function. This should already be C4. Underneath you can specify
the type of problem youre solving. You will be required to solve both minimization and
maximization problems in this assignment, so we select either Min or Max.
In By changing cells you specify the cells the solver is allowed to change. In this example that is C5
and C6. Click the arrow icon at the far right of the box and select these cells. Uncheck the box saying
Make Unconstrained Variables Non-Negative as we do not want to limit the solution to positive
values. If you want to impose some side constraints on your problem, i.e. require that some cell in
your worksheet takes a specific value for the solution to be valid, you may specify these by clicking
the Add-button next to the Subject to the Constraints box. Here you may specify that the cell with
Assignment Instructions

Page 2 of 5

the X-value, C5, must take the same value as a constraint you specify in cell C7, e.g. 0.5. You do not
have to worry about solution methods. Just click OK. The solver will now attempt to find the values
of X and Y that maximizes f, while keeping X equal to the value you entered in C7, and enter them in
cells C5 and C6. A dialogue box may appear informing you that the solver is done. If that happens,
select Keep Solver Solution and click <OK>. There is no need to generate any sensitivity report.

Assignment
Your aim is to construct the efficient frontier, the optimal risky portfolio and the optimal complete
portfolio. You are then asked to provide a brief report, maximum 800 words, for an investment
committee summarising your recommended portfolio. For the report assume that Short sales are
permitted.
In this assignment youll solve the portfolio choice problem in an investment universe with six risky
assets and one risk-free asset. The risky assets include ETFs. For information on ETFs please refer to
the course text BKM section 4.6 and the following ASX link: http://www.asx.com.au/products/etfand-other-etp.htm .
Equities
SPY (SPDR S&P 500 ETF) Representing large cap stocks
IWM (iShares Russell 2000 Index Fund) Representing small cap stocks
VWO (Vanguard Emerging Markets ETF) Representing emerging markets
EFA (iShares MSCI EAFE Index Fund) Representing European, Australian, Asian, and Far
Eastern stocks
Bonds
LQD (iShares iBoxx $ Invest Grade Corp Bond) Representing public, investment-grade,
taxable, fixed income securities
HYG (iShares iBoxx $ High Yield Corporate Bd) Representing public, non-investment-grade,
taxable, fixed income securities
Risk Free Rate
The yield on the 1 year US treasury bill

Analysis
Each problem below forms a part in this overarching exercise. Throughout, you may assume that
there are 252 trading days in a year.
1. Your first step is to source some of the data for your analysis. The return data is sourced
from Yahoo Finance: https://au.finance.yahoo.com/?p=finance.yahoo.com . The Yahoo
historic report provides for the daily series provides the: Date, Open, High, Low, Close,
Volume and Adj Close. The Prices 31Dec08-2Jan15 sheet tab already contains adjusted
returns data for the relevant assets. Now download the unadjusted closing price from Yahoo
Assignment Instructions

Page 3 of 5

Finance for SPY and enter it into the Prices sheet in the appropriate column. Now go to the
Returns sheet and calculate the returns.
2. In the Questions tab in cells D3 to K3 enter functions that estimate the expected daily return
of all assets. Hint: The appropriate estimator for the mean of a distribution is the sample
1

mean: =
=1 . Time series of historic returns are available in the Daily returns sheet.
The excel-function AVERAGE is useful for such calculations.
3. In cells D4 to K4 enter functions that annualize your daily estimates of the expected returns,
i.e., express them on an annual basis. Hint: You are trying to find the return you would have
if you got your estimated daily return every trading day of the year, i.e. 252 days in a row.
4. In cells D5 to J5 enter functions that estimate the (annual) excess return of all assets using
your values in cells D4 to J5. Hint: To annualize excess returns you must take the difference
between the annualized returns, not annualize their difference.
5. In cells D6 to J6 enter functions that estimate the daily variance of returns for all assets.
Hint: Remember that you are estimating the variance from a random sample rather than
from the entire population. The appropriate estimator for the variance of a distribution from
sample data is: =
6.

7.
8.
9.

10.

11.

12.

1
1

=1( ) . Time series of historic returns are available in the

Daily returns sheet. The excel-function VAR.S is useful for such calculations.
In cells D7 to J7 enter functions that annualize your estimates of the daily variance of returns
for all assets. You may assume that daily returns are independent. That is a standard
assumption that is appropriate in an efficient market, which we will discuss in a coming
lecture. It turns out that this calculation is cumbersome. A standard (and reasonable)
approximation for the annualized variance is to multiply the daily variance with the number
of trading days, i.e. 252. This method would be exact if returns were compounded
continuously, rather than on a daily basis as in our data.
In cells D8 to J8 enter functions for the annualized standard deviation of returns. From here
on we will only be concerned with the annualized variables.
In cells D9 to J9 enter functions for the annual Sharpe-ratio for all ETFs.
Having estimated the statistical properties of the assets you can invest in, it is now time to
combine them into some interesting portfolios. Youll do this in the area labelled Portfolio
lab, i.e., the area C12:J17. Enter functions in cells D14 to D17 that calculate the properties
of the portfolio defined by the weights in cells D13 to I13.
Find the efficient frontier by finding portfolio weights that minimize the portfolio variance
for each of the given portfolio returns in cells C24 to C29. Use the Solver and the portfolio
lab to solve the six associated constrained minimization problems. Copy each solution to the
corresponding cells in columns D to I. Also make sure to copy the resulting standard
deviation to the corresponding cell in column J (and take care to copy the value as opposed
to the formula). Hint: the paste area is a convenient place to take the information from.
Now use the Solver and the portfolio lab to find the optimal risky portfolio, , in this
market. Copy the optimal portfolio weights to cells D33 to I33 and the associated portfolio
properties to cells D34 to D37. The efficient frontier will automatically plot in the graph.
Up to this stage we have assumed that Short sales are permitted. Now assume that your
fund has a 130-30 restriction. That is the portfolio mandate requires that the weight for any
ETF has to between the range (130%, -30%). Add this as a constraint in solver and find the

Assignment Instructions

Page 4 of 5

130-30 constrained optimal portfolio with this constraint. Copy the portfolio weights to
cells D41to I41 and the associated portfolio properties to cells D42 to D45.
13. Now assume that Short sales are not permitted. Add this as a constraint in solver and find
the no short optimal portfolio. Copy the portfolio weights to cells D49 to I49 and the
associated portfolio properties to cells D50 to D53.
14. Find the optimal complete portfolio. Assuming that you have preferences represented by
1
2

the utility function U = E(r) A2 and a risk aversion parameter A = 50, enter this value
in D58 and a function in cell C61 that calculates the fraction of your wealth that you should
invest in the optimal risky portfolio. Enter the resulting portfolio properties in cells C62 and
C63, and enter your resulting utility level in cell C64. Note that the utility function used here
assumes that you express E(r) and in decimal form, e.g. 0.2 for 20%.
15. Solve the same problem of finding the optimal complete portfolio again, but use a numerical
solution method this time. That is, specify appropriate formulas in cells E62 to E64 and use
the Solver to find the optimal value in cell E61.

Report
Provide a brief report on the proposed portfolio (maximum 800 words in total). You may wish to
address some of the following points.
We use historical data to forecast the expected ETF returns. Is this a reasonable assumption?
Hint: try deleting a block of say half the earlier years of data and examine how the mean and
standard deviation changes.
How do the various ETFs assist in diversifying portfolio risk and what role does this suggest
about including the respective ETF in the portfolio? Hint: Examine the covariance matrix. Do
any ETF have especially low levels of covariance?
Key assumptions to be highlighted to the investment committee.

Assignment Instructions

Page 5 of 5

You might also like