Week 8 – Chapter 12 & 13
BY SISI ZHANG, [Link].
Learning Objectives
1. Return
2. Risk
3. Portfolio Construction
4. Portfolio Risk
5. Portfolio Optimization
Efficient Market Hypothesis (EMH)
Modern Portfolio Theory
Return
Holding Period Return
𝑷𝑷𝟎𝟎−𝑷𝑷𝟏𝟏 +𝑰𝑰𝟏𝟏
R= 𝑷𝑷𝟎𝟎
You bought a bond at $100, you sold the bond at $110, and at the end of the
period you also received a coupon payment of $5
𝑃𝑃0−𝑃𝑃1 +𝐼𝐼1 $110−$100 +$5 $15
R= = = = 15% 0𝑅𝑅 0.15
𝑃𝑃0 $100 $100
Return
When assets have returns for multiple holding periods, it is necessary to
aggregate those returns into one overall return for ease of comparison and
understanding.
�𝒊𝒊 = 𝑹𝑹𝒊𝒊𝒊𝒊+𝑹𝑹𝑹𝑹𝟐𝟐+⋯+𝑹𝑹𝑹𝑹,𝑻𝑻−𝟏𝟏+𝑹𝑹𝑹𝑹𝑹𝑹 ( For asset i, held for T periods, what is the average
𝑹𝑹
𝑻𝑻
return)
Stock i’s returns for 2019, 2020, 2021 were -50%, 35%, +27%. What is the
arithmetic mean return? (-50%+35%+27%/3=4%)
Return
If we are given the probabilities of an asset’s returns, the expected returns:
E(R) = ∑𝑁𝑁
𝑖𝑖=1 𝑅𝑅𝑖𝑖 xP𝑖𝑖
E(RDWC) = 50%X(-10%) + 50%X(40%) = 15%
E(RGNX) = 50%X20%+50%X5% =12.5%
Return
Geometric Mean Return
Let’s take a look at the previous example where Stock i’s returns for 2019, 2020,
2021 were -50%, 35%, +27% respectively. Is the arithmetic mean return (4% per
year) a good representation of last three years return?
You invested $10,000 in the beginning of 2019 and sold it in the end of 2022.
$10,000 x (1-50%) = $5,000 end of 2019
$5,000 x (1+35%) = $6,750 end of 2020
$6,750 x (1+27%) = $8,572.5 end of 2021
Therefore, you have lost $8,872.5 - $10,000 = -$1,427.5. Yet, your average return
for the last three years is +4%?
Return
It turns out that we need a better way for calculating the average than the
arithmetic mean. We call it the Geometric Mean Return
𝑇𝑇
𝑅𝑅𝐺𝐺𝐺𝐺 = 1 + 𝑅𝑅𝑅𝑅1 𝑋𝑋 1 + 𝑅𝑅𝑅𝑅2 X … X(1 + Ri, T − 1)X(1+Ri, T)-1
Let try the previous example again, we have (1+ (-50%)) for 2019, (1+35%)
for 2020, and (1+27%) for 2021.
3
Therefore, the GMR = 0.5x1.35x1.27 − 1 = -5%
Risk
Standard Deviation (𝝈𝝈) 𝒂𝒂𝒂𝒂𝒂𝒂 Variance (𝝈𝝈2)
In finance we use these two measures to describe the risk of the
securities or portfolio
A standard deviation (or σ) is a measure of how dispersed the data is
in relation to the mean. Low standard deviation means data are
clustered around the mean, and high standard deviation indicates
data are more spread out.
Risk
Normal Distribution
Risk
Mean = Medium = Mode
In addition, mean = Medium = Mode for a normal distribution (Bell CURVE)
For simplicity purposes, we first assume that asset returns follow a normal
distribution
Therefore, we can use Standard Deviation (𝜎𝜎) to describe an assert’s
volatility or riskiness. Things can be different.
Risk
Standard Deviation (𝜎𝜎)
𝑖𝑖=1[𝑅𝑅𝑅𝑅−E(R)]
∑𝑁𝑁 2
𝜎𝜎 =
𝑁𝑁−1
µ :Arithmetic Mean of an asset returns
N: Number of observations
Ri: Return of an asset in period i
Example: Stock A’s returns for 2019, 2020, 2021 were -50%, 35%, +27%.
1) What is the arithmetic mean return?
(-50%+35%+27%/3=4%)
2) What is Stock A’s 𝜎𝜎? (Next Page)
Risk
Standard Deviation (𝜎𝜎)
Mean = 4%
−50%−4% 2+ 35%−4% 2+ 27%−4% 2
𝜎𝜎 = = 0.4694 =46.94%
3−1
𝜎𝜎2= (46.94%)2= 0.2203
Risk
Standard Deviation (𝜎𝜎)
Mean = 4%
−50%−4% 2+ 35%−4% 2+ 27%−4% 2
𝜎𝜎 = = 0.4694 =46.94%
3−1
𝜎𝜎2= (46.94%)2= 0.2203
Risk
What if we are given probabilities of returns for each asset
𝜎𝜎2 = ∑𝑁𝑁
𝑖𝑖=1[𝑅𝑅𝑅𝑅 − 𝐸𝐸(𝑅𝑅)] xP𝑖𝑖
2
Risk
𝜎𝜎2(RABA) = ?
First, we need to calculate the mean or average return
E(R) = ∑𝑁𝑁
𝑖𝑖=1 𝑅𝑅𝑖𝑖 xP𝑖𝑖 = -20%x25%+-30%x30%+5%x45% = 6.25%
Second, we need to calculate the sum of squares
𝜎𝜎2 = ∑𝑁𝑁
𝑖𝑖=1[𝑅𝑅𝑅𝑅 − 𝐸𝐸(𝑅𝑅)]2P𝑖𝑖 = (-20% - 6.25%) x25%+ (30% - 6.25) x 30% +
2 2
(5% - 6.25%)2x 45% = 0.0172+0.0169+0.00007 = 0.03422
𝜎𝜎 = 𝜎𝜎2 = 0.03422 = 0.185 =18.5%
Risk
𝜎𝜎2(RABA) = ?
First, we need to calculate the mean or average return
E(R) = ∑𝑁𝑁
𝑖𝑖=1 𝑅𝑅𝑖𝑖 xP𝑖𝑖 = -20%x25%+-30%x30%+5%x45% = 6.25%
Second, we need to calculate the sum of squares
𝜎𝜎2 = ∑𝑁𝑁
𝑖𝑖=1[𝑅𝑅𝑅𝑅 − 𝐸𝐸(𝑅𝑅)]2P𝑖𝑖 = (-20% - 6.25%) x25%+ (30% - 6.25) x 30% +
2 2
(5% - 6.25%)2x 45% = 0.0172+0.0169+0.00007 = 0.03422
𝜎𝜎 = 𝜎𝜎2 = 0.03422 = 0.185 =18.5%
Risk
Skewness
Risk
Kurtosis
Portfolio Construction
Portfolio Return
when we combine different securities into one portfolio
RP= ∑𝑁𝑁 𝑁𝑁
𝑖𝑖=1 𝑤𝑤𝑖𝑖 R𝑖𝑖and ∑𝑖𝑖=1 𝑤𝑤𝑖𝑖= 1
An Example, you have $100,000 to invest. You are ready to put 60% into
asset A with an average return of 8% and the reminder into assert B with
an average return of 6.5%. What is your portfolio’s expected return
RP= 60%x8% + (1-60%)x6.5% = 4.8%+2.6% = 7.4%
Portfolio Construction
Portfolio Return
when we combine different securities into one portfolio
RP= ∑𝑁𝑁 𝑁𝑁
𝑖𝑖=1 𝑤𝑤𝑖𝑖 R𝑖𝑖and ∑𝑖𝑖=1 𝑤𝑤𝑖𝑖= 1
An Example, you have $100,000 to invest. You are ready to put 60% into
asset A with an average return of 8% and the reminder into assert B with
an average return of 6.5%. What is your portfolio’s expected return
RP= 60%x8% + (1-60%)x6.5% = 4.8%+2.6% = 7.4%
Portfolio Construction
Portfolio Risk
We need to calculate the co-movements of securities within a portfolio
𝜎𝜎2(p) = ∑𝑁𝑁
𝑖𝑖,𝑗𝑗=1 𝑤𝑤𝑖𝑖 , 𝑤𝑤𝑗𝑗 Cov (Ri,Rj)
For a two securities portfolio we can re-write the above formula below
𝜎𝜎2(p)= 𝑤𝑤12 𝜎𝜎12 + 𝑤𝑤22 𝜎𝜎22 + 2w1w2 Cov( R1,R2)
𝜎𝜎(p)= 𝑤𝑤12 𝜎𝜎12 + 𝑤𝑤22 𝜎𝜎22 + 2w1w2 Cov( R1,R2)
Portfolio Construction
Covariance and Correlation Coefficient
Covariance Cov (Ri,Rj) or 𝜎𝜎ij : Measure the joint variability of two random
variables
Positive means that the pair tend to move in the same direction whereas
negative means they tend to move in the opposite directions in general.
Calculating Covariance Cov (Ri,Rj) is beyond the scope of the course, and
you are given the number during exam if you asked.
Cov (Ri,Rj) = ρi,j 𝜎𝜎i 𝜎𝜎j (ρ is Correlation Coefficient)
Therefore, 𝜎𝜎2(p)= 𝑤𝑤12 𝜎𝜎12 + 𝑤𝑤22 𝜎𝜎22 + 2w1w2 ρi,j 𝜎𝜎i 𝜎𝜎j
Portfolio Construction
Correlation Coefficient
1. ρ =1, Returns of the two assets are perfectly positively
correlated. Assets i & j move together 100 percent of the
time
2. ρ = -1, Returns of the two assets are perfectly negatively
correlated. Assets i & j move in opposite directions 100
percent of the time
3. ρ = 0, Returns of the two assets are uncorrelated. Movement
of Asset i provides no prediction regarding the movement of
Asset j.
4. ρ is always between -1 and +1 for any pair.
Portfolio Construction
Portfolio of Two Assets
Assume you are a UK investor holding a portfolio invested 60% in UK
large-capitalization equities (the FTSE 100 Index) and 40% in local
medium-duration Treasury bonds (“gilts”). The expected return on
the FTSE 100 is 5.5% and on the medium-duration gilts it is 0.7%. The
risk (standard deviation of returns) is 13.2% and 4.2%, respectively.
The correlation coefficient of the pair is –0.01
Portfolio Construction
Portfolio of Two Assets
Assume you are a UK investor holding a portfolio invested 60% in UK large-
capitalization equities (the FTSE 100 Index) and 40% in local medium-duration
Treasury bonds (“gilts”). The expected return on the FTSE 100 is 5.5% and on
the medium-duration gilts it is 0.7%. The risk (standard deviation of returns) is
13.2% and 4.2%, respectively. The correlation coefficient of the pair is –0.01
RP= ∑𝑁𝑁
𝑖𝑖=1 𝑤𝑤𝑖𝑖 R𝑖𝑖 = 0.6 X 5.5% + (1-0.6) X 0.7% = 3.6%
𝜎𝜎2(p)= 𝑤𝑤12 𝜎𝜎12 + 𝑤𝑤22 𝜎𝜎22 + 2w1w2 ρi,j 𝜎𝜎i 𝜎𝜎j
= [(0.62𝑋𝑋𝑋.13222) + 0.42𝑋𝑋𝑋.04222 +
2𝑋𝑋𝑋.6𝑋𝑋𝑋.4𝑋𝑋(−0.01)𝑋𝑋𝑋.132𝑋𝑋𝑋.042]
𝜎𝜎(p)= 8.1%
Portfolio Construction
Introduction of Risk-Free Asset
Imagine a world with two assets: a risky asset and a risk-free asset to be used to
construct a portfolio
A Risk-free asset’s return has no variability (High interest saving account)
E(Rp)=W1Rf+(1−W1)E(Ri)
𝜎𝜎2(p)= 𝑊𝑊12 𝜎𝜎𝑓𝑓2 + (1-W1)2𝜎𝜎𝑖𝑖2 + 2W1(1-W1) ρi,f 𝜎𝜎i 𝜎𝜎f = (1-W1)2𝜎𝜎𝑖𝑖2 = (1-W1)2𝜎𝜎𝑖𝑖2
𝜎𝜎(p) = (1-W1) 𝜎𝜎i
Portfolio Construction
Capital Allocation Line
(E(Ri)−R𝑓𝑓) 𝜎𝜎p :denotes for portfolio’s Standard Deviation
E(Rp)= Rf+ 𝜎𝜎p
𝜎𝜎i E(Ri); expected return of asset 𝑖𝑖
𝜎𝜎i: Asset i′ s standard Deviation
E(Ri) − R𝑓𝑓:Risk Premium
Portfolio Construction
Market Risk
Diversification Effect, we saw the two assets portfolio previously. If we try
to incorporate more and more assets into the portfolio, we are able to
reduce the risk for the portfolio for a given required return (Do not put all
your eggs in one basket)
Systematic Risk, it is the level of risk that cannot be reduced for a given
required return by diversification or risk investors required to assume and
cannot be diversified away
Unsystematic Risk, these are risks specific to an individual security or an
industry but not to the entire market
Portfolio Construction
Systematic Vs. Unsystematic Risk
Interest rates, inflation, economic cycles, political uncertainty, and
widespread natural disasters.
Examples of nonsystematic risk could include the failure of a drug trial
or an airliner crash. All these events will directly affect their respective
companies and possibly industries but have no effect on assets that are
far removed from these industries. Investors can avoid nonsystematic
risk through diversification by forming a portfolio of assets that are not
highly correlated with one another.
Portfolio Construction
Systematic Vs. Unsystematic Risk
Portfolio Optimization
Efficient Market Hypothesis (EMH)
The efficient market hypothesis (EMH), alternatively known as the efficient
market theory, is a hypothesis that states that share prices reflect all
available information and consistent Alpha generation is impossible.
Alpha:
Alpha (α) is a term used in investing to describe an investment
strategy’s ability to beat the market.
Alpha is also often referred to as excess return or the abnormal rate of
return in relation to a benchmark, when adjusted for risk.
Portfolio Optimization
Efficient Frontier
As you include more and more assets into your portfolio, some are weakly correlated
and some are negatively correlated, you expand your Investment opportunity set: along
the frontier, for a given required return, you can find a portfolio with the lowest risk.
Portfolio Optimization
Efficient Frontier
A rational, risk-averse investor would only choose a portfolio from the
efficient frontier since this portfolio is lowest risk portfolio for the required
return
Another observation is that as we move along the curve from left to right,
expected return is increasing for every additional unit of risk, however, the
increase in return is decreasing.
You can take all the risk you want, you may not increase your potential
return much
Portfolio Optimization
Risk Free Asset and CML
Portfolio Optimization
Mathematical Expression of the CML
After we combine all risky assets and a risk-free asset. We can obtain the following
mathematical expression
𝐸𝐸 𝑅𝑅𝑇𝑇 −Rf
E(Rp) = Rf+ ( ) X σp E(Rp) = Wf Rf + (1- Wf) 𝐸𝐸 𝑅𝑅𝑇𝑇
σ𝑇𝑇
E(Rp): the expected return of a diversified portfolio
Rf : the expected return of your risk-free asset
σ𝑇𝑇: the tangent portfolio’s standard deviation
σp: your diversified portfolio’s standard deviation
Wf: percentage allocated towards risk free asset
Since we have named the tangent portfolio as the market portfolio, we can replace T with
M
𝑬𝑬 𝑹𝑹𝑻𝑻 − Rf: Market risk Premium
Portfolio Optimization
Example 1
Mr. Miles is a first-time investor and wants to build a portfolio using only
US T-bills and an index fund that closely tracks the S&P 500 Index. The T-
bills have a return of 5 percent. The S&P 500 has a standard deviation of
20 percent and an expected return of 15 percent. If Mr. Miles’s goal is to
achieve 8% return, how much risk he has to assume and what is the
portion invested into the market portfolio?
Portfolio Optimization
Example 1
Mr. Miles is a first-time investor and wants to build a portfolio using only
US T-bills and an index fund that closely tracks the S&P 500 Index. The T-
bills have a return of 5 percent. The S&P 500 has a standard deviation of
20 percent and an expected return of 15 percent. If Mr. Miles’s goal is to
achieve 8% return, how much risk he has to assume and what is the
portion invested into the market portfolio?
Portfolio Optimization
Solutions
𝐸𝐸 𝑅𝑅𝑇𝑇 −Rf
E(Rp) = Rf+ ( ) X σp
σ𝑇𝑇
8% = 5% + σp x(15% - 5%)/20%
σp = 6%
E(Rp) = Wf Rf + (1- Wf) 𝐸𝐸 𝑅𝑅𝑇𝑇
8% = Wf x5% + (1- Wf )x15%
Wf= 70%, So WM= 30%
Portfolio Optimization
What is the risk-free asset’s return changed to 6%, please recalculate how
much risk he has to assume in his portfolio and what is the portion should be
invested into the market portfolio?
𝐸𝐸 𝑅𝑅𝑇𝑇 −Rf
E(Rp) = Rf+ ( X σp)
σ𝑇𝑇
8% = 6% + σp x(15% - 6%)/20%
σp = 4.4%
E(Rp) = Wf Rf + (1- Wf) 𝐸𝐸 𝑅𝑅𝑇𝑇
8% = Wf x6% + (1- Wf )x15%
Wf= 78%, WM= 22%
Portfolio Optimization
Capital Asset Pricing Model (CAPM )
The capital asset pricing model, or CAPM, is a financial model that calculates the
expected rate of return for an asset or investment.
Portfolio Optimization
Beta
Portfolio Optimization
Beta coefficient (β): This is a metric which informs the investor on
how much systematic risk an asset/investment has relative to
average. For example, if negative news hit the markets today, the
stocks with betas over 1.0 would theoretically fall more than
average while the stocks with betas under 1.0 would not fall as
much as the average.
β can take any value, by definition Beta for the market portfolio is 1
For a portfolio of securities, we can add them linearly we obtain the
portfolio Beta
βp= ∑𝑛𝑛𝑖𝑖=1 𝑤𝑤𝑖𝑖βi; ∑𝑛𝑛𝑖𝑖=1 𝑤𝑤𝑖𝑖 = 1
Portfolio Optimization
Security Market Line SML
The Security Market Line (SML) is a visualization of the Capital Asset Pricing Model (CAPM)
and shows the theoretical relationship between risk and return between securities and the
entire market.
Portfolio Optimization
Beta calculations
Mr. Miles is a first-time investor and wants to build a portfolio individual stocks. After
some planning, he decided to invest into 5 stocks from TSX index. Expected return for the
TSX index is 10%, and the 10-year Canadian Government bond yields 5%. What is his
expected return based on CAPM?
Βp= 20%x1.05+33%x2.58+25%x0.86+22%x3.97=2.1498
E(Ri) = Rf+ βi (𝐸𝐸 𝑅𝑅𝑇𝑇 − Rf) = 5% + 2.1498 x(10%-5%) = 15.749%
Portfolio Optimization
Comparing Portfolios
Sharp Ratio
E(Rp) − Rf
SR =
σ𝑝𝑝
Portfolio’s risk premium per unit of risk
The Treynor Ratio
E(Rp) − Rf
TR = β𝑝𝑝
Portfolio’s risk premium per unit of systematic risk
Portfolio Optimization
Jensen’s Alpha
Since we can calculate the portfolio’s expected return based on the
portfolio’s systematic risk factor (Beta), we then calculate Alpha
the difference
Alpha = RP- RP(based on CAPM Model)
Thank you!