Investment Portfolio & Risk Analysis
Investment Portfolio & Risk Analysis
➢ Sharpe Ratio
FINS5513
Portfolio Basics
FINS5513 4
What is a Portfolio?
❑ A combination of multiple assets and/or securities owned by an investor
➢ The aim of owning multiple assets is to achieve diversification
5
The Investment Process
❑ The process of portfolio construction can be undertaken top-down or bottom-up
❑ Top-down portfolio construction
➢ Asset allocation – choosing between broad asset classes and determining what
proportions of the portfolio should be invested in each asset class
➢ Top-down starts with asset allocation, then we decide which securities to hold in each asset
class Video 2AV1: “Investopedia: Strategic Asset Allocation”
❑ Bottom-up portfolio construction
➢ Security selection – choosing which individual securities to hold within each asset class
➢ Construct portfolios from securities that are attractively priced with less concern for the
resulting asset allocation
➢ May lead to being overweight or underweight certain sectors or security types
Further Reading 2AR1: Various studies
on asset allocation vs security selection
6
Themes in Portfolio Management
FINS5513 7
Key Concepts
❑ Asset/fund managers are called the “buy-side” – they buy the services of sell-side firms such
as broker/dealers who sell securities and provide research/recommendations to the buy-side
➢ A number of important buy-side themes will be explored
❑ Efficient markets
➢ Efficient markets price securities quickly and accurately incorporating all relevant
information
➢ Are markets efficient in practice?
❑ Risk-return trade-off
➢ In efficient and competitive markets higher expected return should come with higher risk
8
Active vs Passive Management
❑ Active vs Passive management
➢ Active management – attempting to improve performance by identifying mispriced
securities (through security analysis and security selection)
• Active managers attempt to outperform a prescribed market benchmark such as the
S&P500 or the ASX200
➢ Passive management – holding diversified portfolios (little time spent on security selection)
• Passive managers attempt to track a prescribed market benchmark such as the S&P500
or the ASX200
➢ If markets are efficient, why bother with active management?
• If an investor were to believe in efficient markets, the only important decision is asset
allocation, not security selection
❑ There is a significant evidence that the majority of active fund managers underperform their
benchmarks, and overall returns from actively managed funds lag wider stock indices
Further Reading 2AR2: “SPIVA 2020 Scorecards”
Video 2AV2: “Interview with Andrew Innes S&P on active vs passive performance” 9
Other Key Themes
❑ Traditional vs Alternative
➢ Traditional – long-only, unleveraged funds focused on equity, fixed income and/or balanced
(multi-asset) asset classes
• Charge management fees based on FUM
➢ Alternative – hedge funds, private equity, venture capital - often leveraged
• Charge management fees based on FUM and performance fees (or “carried interest”)
❑ Growth vs Value
➢ Growth – focuses on early stage emerging companies whose growth is expected to
significantly outperform wider industry trends. Often follows momentum and trends
➢ Value – concentrates on stocks that appear to be trading for less that their intrinsic value.
Focuses on low P/E, low Price/Book, and high free cash flow stocks
➢ Traditionally, value stocks have provided higher returns than growth, however this trend
appears to have reversed post GFC
Further Reading 2AR3: “Where’s the value in value investing?”
Video 2AV3: “Howard Marks’s thoughts on value vs growth investing” 10
Recent Trends
❑ Increase in passive investing due to lower cost and underperformance of active managers
❑ Increased variety and specialisation in ETFs eg thematic and factor based ETFs
❑ Increased use of high frequency trading and other quant methods using advanced statistical
and programming based techniques
➢ Attempt to take advantage of very short-term anomalies in the market
❑ Wider use of new data sources such as: social media; imagery and sensor data (customer
tracking, carpark monitoring, weather conditions etc); management psychological studies etc
to guide investing
❑ Robo-advisors and other algorithm-driven financial planning digital platforms (with no or little
human involvement)
11
2.2 Measuring Return and
Risk
FINS5513
Return Measurement
FINS5513 13
Holding Period Return
❑ Investor returns from holding an asset come from two basic sources:
➢ Income received periodically such as interest (debt security) and dividends (equity security)
➢ Capital gains (or losses) from the price of the asset increasing (decreasing)
❑ Holding Period Return (HPR) is the return on an asset during the period it is held
➢ The holding period ends when the asset is sold or matures/expires (for finite life assets)
Capital gain Income
component 𝑃𝑇 − 𝑃0 + 𝐼𝑇 component
𝐻𝑃𝑅 =
𝑃0
𝑃0 = Price at the beginning of period T
𝑃𝑇 = Price at the end of period T
𝐼𝑇 = Total income received over the holding period T (eg interest, coupons, dividends)
❑ Example 2A1: You buy a share for $75 and sell it 9 months later for $84. It paid a div of $2.25:
➢ P0 = 75 PT = 84 IT = 2.25
84 −75+2.25
𝐻𝑃𝑅 = = 0.15 = 𝟏𝟓%
75
14
APR and EAR
❑ The HPR gives the total return over the holding period without regard to the time period
❑ We can annualise the HPR in two ways:
➢ Assuming simple interest – we call this the Annualised Percentage Rate (APR)
➢ Assuming compound interest – we call this the Effective Annual Rate (EAR)
❑ Assume: T = holding period expressed in years (eg T=2 for 2-year hold; T=0.25 for 3 months;
T=5.5 for 5 years and 6 months)
❑ Annualized percentage rate (APR)
𝐻𝑃𝑅
𝐴𝑃𝑅 =
𝑇
❑ Effective Annual Rate (EAR)
1/T
𝐸𝐴𝑅 = 1 + 𝐻𝑃𝑅 -1
15
APR and EAR
❑ The EAR accounts for compounding interest, not just simple interest (ie “interest on interest”)
➢ If compounding is annual: EAR = APR (at year end)
➢ If compounding is more frequent then annual: EAR > APR (at year end)
17
Measuring Expected Return
FINS5513 18
Expected Return: Ex-Ante
❑ The reward from an investment is its return
➢ Since returns are generally uncertain (or “stochastic”) we deal with expected returns
19
Expected Return: Ex-Ante
❑ Example 2A3: After extensive simulations, Quant Fund has determined that the distribution
of returns for Walmart (WMT) in different probability weighted economic future scenarios is
given by:
Economic Scenario Scenario Probability Scenario Return
Boom 0.25 38.0%
Growth 0.50 14.0%
Flat 0.20 -7.5%
Recession 0.05 -32.0%
Excel 2AE2: “2A – Calculating Ex-Ante & Ex-Post ER, Var & SD”
20
Expected Return: Ex-Post
❑ Estimating expected returns by projecting future scenarios can have a high level of
forecasting error
❑ Therefore, expected return is often estimated using the average (or mean) historical
(backward-looking or ex-post) sample rates of return, denoted 𝒓ത by using the formula:
1
𝑟ҧ = σ𝑛𝑡=1 𝑟𝑡
𝑛
𝑟𝑡 = Return at time t
❑ Example 2A4: Quant Fund analysed 10-year historical returns for WMT as shown:
2020 2019 2018 2017 2016 2015 2014 2013 2012 2011
WMT Returns 23.3% 30.2% -3.4% 46.5% 16.0% -26.6% 11.9% 18.2% 17.0% 13.9%
➢ The expected return can be estimated from the historical average return 𝑟ҧ :
𝑟ҧ = (.233 + .302 −.034 + .465 + .16 − .266 + .119 + .182 + .17 + .139) / 10
𝑟ҧ = .147 or 14.70%
21
Measuring Risk
FINS5513 22
What is Risk?
❑ We seek to maximise return because return maximises wealth
➢ However, we seek return in a world of uncertainty
❑ In finance, risk refers to the possibility that realised outcomes differ (better or worse) from
expectation
➢ We seek not to avoid risk, but to incorporate it appropriately into decision making
❑ Think of return as the “reward” and risk as the “cost” of that reward
❑ So, how do we measure risk?
➢ In a quantitative sense, risk is a measure of the volatility of our returns
Video 2AV4: RB: “Why is risk - measured by volatility - a problem for fund managers?”
23
Risk: Ex-Ante
❑ Volatility is the sum of total (squared) deviations from our expectations
➢ This is known as the variance which on an ex-ante basis is given by:
𝜎 2 = σ𝑠 𝑝 𝑠 [𝑟 𝑠 − 𝐸 𝑟 ]2
➢ To return to original units (rather than squares), we use the Standard Deviation:
𝜎= 𝜎2
❑ Example 2A5: Determine the standard deviation for Quant Fund’s ex-ante analysis of WMT
➢ Step #1 – Derive 𝐸(𝑟) = 13.40%
➢ Step #2 – Take the actual return in each scenario and subtract 𝐸(𝑟)
➢ Step #4 – Multiply each scenario’s squared differences by its probability and sum them:
𝜎 2 = .25(.38 −.134)2 + .50(.14 −.134)2 + .20(− .075 −.134)2 + .05(−.32 −.134)2 = 0.034
➢ Step #5 – Standard deviation is the square root of the variance: 𝜎 = 0.034 = 𝟏𝟖. 𝟒𝟗%
24
Risk: Ex-Post
❑ We can also use historical (ex-post) data to estimate the risk
❑ When conducting ex-post (backward-looking) analysis, each historical data point is
considered equally likely and therefore we do not probability weight them. However, we
divide by n – 1 (rather than n) to account for estimation error as 𝒓ത is only an estimation of 𝐸(𝑟)
❑ The unbiased standard deviation estimate 𝜎ො is given by:
2
σ𝑛
𝑡=1 𝑟𝑡 − 𝑟ҧ
𝜎ො =
𝑛−1
❑ Example 2A6: Determine the standard deviation for Quant Fund’s ex-post analysis of WMT
ො 2 = [ (.233 - .147)2 + (.302 - .147)2 + (−.034 - .147)2 + (.465 - .147)2 + (.16 - .147)2 +
➢ 𝜎
(−.266 - .147)2 + (.119 - .147)2 + (.182 - .147)2 + (.17 - .147)2 + (.139 - .147)2 ] / 9
= 0.3386
➢ 𝜎ො = 0.3386 = 𝟏𝟗. 𝟒𝟎% Excel 2AE2: “2A – Calculating Ex-Ante & Ex-Post ER, Var & SD”
25
Sharpe Ratio
FINS5513 26
Reward to Volatility (Sharpe) Ratio
❑ Now that we have quantified return and risk individually, how do we relate the risk/reward
relationship in one measure?
➢ Divide return (the “reward”) by risk (the “cost”) and state it as a ratio
❑ We often look at the “excess return” above the risk-free rate, rather than the total return
➢ This is because part of the return can be earned for no risk by investing in a risk-free asset
➢ On a forward looking basis we often refer to the expected excess return above the risk-free
rate for a risky asset as the Risk Premium
❑ The Sharpe Ratio is given by:
𝐸 𝑟𝑖 − 𝑟𝑓
Sharpe ratio for security i : 𝑆𝑖 =
𝜎𝑖
𝑟𝑓 = risk-free rate
𝐸 𝑟𝑖 − 𝑟𝑓 = Risk premium for security i
𝜎𝑖 = Standard deviation of excess returns for security i
27
The Importance of the Sharpe Ratio
❑ The Sharpe ratio measures return per unit of risk. The higher the Sharpe ratio - the higher
the incremental return received per unit of risk
➢ In other words, the higher the Sharpe ratio the better (the more attractive the investment)
➢ As the Sharpe ratio is straight forward to calculate and easy to interpret, it is one of the most
widely used appraisal measures for assessing risk against reward
• However, within a portfolio context it does have limitations which we will explore later
❑ Example 2A7: Determine the Sharpe ratio for both Quant Fund’s ex-ante and ex-post
analysis of WMT. Assume a risk-free rate of 3.0%
.1340 − .03
➢ Ex-ante: SWMT = = 0.562
.1849
.1470 − .03
➢ Ex-post: 𝑆መ WMT = = 0.603
.1940
FINS5513
Distribution of Returns
FINS5513 30
Which Fund is Preferred?
❑ Consider two funds: All Weather (AW) and Traditional Portfolio (TP)
➢ AW and TP have the same expected return 𝐸(𝑟) of 10%, but which would you choose?
Probab ility
0.2
s AW =5%
0.15
AW A
0.1
TPB
0.05
s TP =10%
0
Retu rn (%)
0 5 10 15 20
32
Normal Distributions
❑ Investment management is simplified when returns (which are uncertain) are approximated
as a normal distribution:
➢ Normal distribution assumes returns are symmetric around the mean
➢ Future return probabilities can be estimated using only mean and standard deviation
❑ If returns are not normally distributed, standard deviation is no longer a complete measure of
risk and we must also consider skewness and kurtosis
Further Reading
Skew Kurtosis 2AR5: “How ‘Tail Risk’
changes over the
market cycle”
Video 2AV5:
“Nassim Taleb - What is
a "Black Swan?”
33
Risk Aversion
FINS5513 34
Mean-Variance Criterion
❑ Mean-variance analysis requires the mean and standard deviation of returns
➢ We graph expected return (y-axis) against standard deviation (x-axis)
❑ Historical market returns show there is a risk premium in the market (indicates risk aversion):
➢ Since 1926, U.S. risk-free assets (1-month T-bills) returned ~3.4% annually while risky
assets (US stocks) returned ~11.7% – resulting in a ~8.3% risk premium with 𝜎 = 20.4%
➢ Market takes additional risk only for a commensurate return – indicating risk aversion
Further Reading 2AR6: “Australian Investor Study 2020” Figures 20, 22, 48, 49, 52-55 36
Preference and Utility
FINS5513 37
Preference and Utility
❑ Utility is a measure of satisfaction / welfare / happiness of an investor
❑ When an investor prefers Asset A over Asset B, we say that Asset A provides the investor with
greater utility
❑ In order to work with preferences mathematically, we use utility functions
➢ A utility function assigns a value to each outcome so that preferred outcomes get higher
utility values
❑ For simplicity, the fundamental assumption in finance is that utility is derived from wealth
➢ We assume that the more money an investor has, the better their ability to achieve
preferred outcomes
➢ Therefore, in finance:
• More is better – maximise return which maximises wealth
• More certainty is better – risk aversion
38
Logarithmic Wealth Utility Function
❑ A common specification of the wealth utility function is U(W) = ln(W)
❑ The logarithmic expression results in a concave function
➢ The concavity indicates that the incremental utility we gain from increases in wealth is less
than the utility we lose from equivalent decreases in wealth
➢ The concavity captures risk aversion – risk averse investors would not take a 50/50 bet
4.5
4 Utility curve
2.5
Utility
2
Average utility from A and B
1.5
0.5
Wealth level A: 1
0
1 11 21 31 41
Wealth
39
Risk-Reward Trade-off
❑ Choosing the preferred asset when one dominates is straight forward
❑ But what about where no asset dominates?
Fund Expected Return Risk Premium Risk
(rf = 5%) σ
Low-Risk 7.0% 2.0% 5.0%
Medium-Risk 9.0% 4.0% 10.0%
High-Risk 13.0% 8.0% 20.0%
❑ In the example, return increases but so does risk (all funds have the same Sharpe ratio)
❑ Each portfolio receives a utility score indicating the investor’s risk/return trade-off
❑ The portfolio with the highest utility score is preferred
40
Utility Function
❑ What is a reasonable method for determining a utility score?
❑ As wealth is dependent on risk and return, we derive a utility function based on risk and return
❑ For investments, we assume a quadratic utility function:
U = E ( r ) − 1 As 2
2
U = Utility
A = Coefficient of risk aversion (a constant)
½ = A scaling factor
➢ Questionnaires
➢ Discussion with broker/advisor
❑ Aggressive investor chooses the High-Risk fund, the others choose the Medium-Risk fund
➢ Risk aversion doesn’t mean the investor doesn’t take risk – rather it means the investor puts
a higher price (return) on taking risk. For example: even the conservative investor does not
pick the Low-Risk fund
43
Indifference Curves
FINS5513 44
Indifference Curves
❑ We can illustrate our preferences through indifference curves
➢ Plotted in the risk-return (𝐸 𝑟 − 𝜎) space that connect points giving equal utility
➢ For example, to draw the indifference curve for U = 10%, choose all asset portfolio
combinations of E(r) and σ which yield a utility score of 10%
➢ Graphical representation of the utility function. Called an Indifference “Curve” because the
utility function is a quadratic equation
➢ Note that two indifference curves with different utility levels never intersect
U = 25%
σ 45
How to Plot an Indifference Curve
❑ Each plot point on an indifference curve represents a risk and return combination which
provides the same utility score
❑ Example 2A9: For an Aggressive investor with risk aversion coefficient A=2, and a
Conservative investor with A=5, plot two indifference curves with U=0.03 and U=0.09
A=5 is
➢ Given a specific value of A, steeper
indifference curves above and to than A=2
the left offer higher utility than lower
curves and don’t intersect Certainty
equivalent Higher Utility
➢ More risk averse investors (A) return.
have steeper indifference curves Plot first.
47