Portfolio Mathematics

LEARNING MODULE 5
LEARNING OUTCOMES

The students should be able to:

u calculate and interpret the expected value, variance, standard deviation,
covariances, and correlations of portfolio returns
u calculate and interpret the covariance and correlation of portfolio returns
using a joint probability function for returns
u define shortfall risk, calculate the safety-first ratio, and identify an optimal
portfolio using Roy’s safety-first criterion
PORTFOLIO EXPECTED RETURN AND
VARIANCE OF RETURN
u The expected return on the portfolio (E(Rp)) is a weighted average of the
expected returns (R1 to Rn) on the component securities using their respective
proportions of the portfolio in currency units as weights (w1 to wn):
𝐸 𝑅! = 𝐸(𝑤" 𝑅" + 𝑤# 𝑅# + ⋯ + 𝑤$ 𝑅$ )
u Accordingly, portfolio variance is as follows:
#
𝜎 # 𝑅! = 𝐸 𝑅! − 𝐸 𝑅!
Exhibit 1: Weights and Expected Returns
of Sample Portfolio

u We calculate the expected return on the portfolio as 11.75 percent:

𝐸(𝑅𝑝) = 0.50(13%) + 0.25(6%) + 0.25(15%) = 11.75%
PORTFOLIO EXPECTED RETURN AND
VARIANCE OF RETURN
u Given two random variables Ri and Rj, the covariance between Ri and Rj is as
follows:
𝐶𝑜𝑣 𝑅% , 𝑅& = 𝐸 (𝑅% −𝐸𝑅% )(𝑅& − 𝐸𝑅& )
u Start with the definition of variance for a three-asset portfolio and see how it
decomposes into three variance terms and six covariance terms.
𝜎 # 𝑅! = 𝐸 𝑤" 𝑅" + 𝑤# 𝑅# + 𝑤' 𝑅' − 𝐸 𝑤" 𝑅" + 𝑤# 𝑅# + 𝑤' 𝑅' # = 𝑤"# 𝜎 # 𝑅" +
𝑤## 𝜎 # 𝑅# + 𝑤'# 𝜎 # 𝑅' + 2𝑤" 𝑤# 𝐶𝑜𝑣(𝑅" , 𝑅# ) + 2𝑤" 𝑤' 𝐶𝑜𝑣(𝑅" , 𝑅' ) ) +
2𝑤# 𝑤' 𝐶𝑜𝑣(𝑅# , 𝑅' )
u Moreover, this expression generalizes for a portfolio of any size n to
$ $

𝜎 # 𝑅! = 0 0 𝑤% 𝑤& 𝐶𝑜𝑣(𝑅% , 𝑅& )

%(" &("
Exhibit 2: Inputs to Portfolio Expected
Return and Variance
Exhibit 3: Covariance Matrix
PORTFOLIO EXPECTED RETURN AND
VARIANCE OF RETURN
u The correlation between two random variables, Ri and Rj, is defined as
follows:
𝑝 𝑅% , 𝑅& = 𝐶𝑜𝑣 𝑅% , 𝑅& / 𝜎(𝑅% )𝜎(𝑅& )
Exhibit 4: Correlation Matrix of Returns
EXAMPLE 1 Portfolio Expected Return and
Variance of Return with Varying Portfolio
Weights
Anna Cintara is constructing different portfolios from the following two stocks:
1. Calculate the covariance between the returns on the two stocks.
2. What is the portfolio expected return and standard deviation if Cintara puts 100
percent of her investment in Stock 1 (w1 = 1.00 and w2 = 0.00)? What is the portfolio
expected return and standard deviation if Cintara puts 100 percent of her investment
in Stock 2 (w1 = 0.00 and w2 = 1.00)?
3. What are the portfolio expected return and standard deviation using the current
portfolio weights?
4. Calculate the expected return and standard deviation of the portfolios when w1
goes from 0.00 to 1.00 in 0.10 increments (and w2 = 1 – w1). Place the results (stock
weights, portfolio expected return, and portfolio standard deviation) in a table, and
then sketch a graph of the results with the standard deviation on the horizontal axis
and expected return on the vertical axis.
Exhibit 5: Description of Two-Stock
Portfolio
EXAMPLE 1 Portfolio Expected Return and
Variance of Return with Varying Portfolio
Weights
EXAMPLE 1 Portfolio Expected Return and
Variance of Return with Varying Portfolio
Weights
QUESTION SET

1. US and Spanish bonds returns measured in the same currency units have
standard deviations of 0.64 and 0.56, respectively. If the correlation between the
two bonds is 0.24, the covariance of returns is closest to:
A. 0.086.
B. 0.335.
C. 0.390.
2. The covariance of returns is positive when the returns on two assets tend to:
A. have the same expected values.
B. be above their expected value at different times.
C. be on the same side of their expected value at the same time.
QUESTION SET

3. Which of the following correlation coefficients indicates the weakest linear

relationship between two variables?
A. –0.67
B. –0.24
C. 0.33
4. An analyst develops the following covariance matrix of returns:
The correlation of returns between the hedge fund and the market index is closest to:

A. 0.005.
B. 0.073.
C. 0.764.
QUESTION SET

5. All else being equal, as the correlation between two assets approaches +1.0,
the diversification benefits:
A. decrease.
B. stay the same.
C. increase.
6. Given a portfolio of five stocks, how many unique covariance terms, excluding
variances, are required to calculate the portfolio return variance?
A. 10
B. 20
C. 25
QUESTION SET

7. Which of the following statements is most accurate? If the covariance of returns between two
assets is 0.0023, then the:
A. assets’ risk is near zero.
B. asset returns are unrelated.
C. asset returns have a positive relationship.
8. A two-stock portfolio includes stocks with the following characteristics:
What is the standard deviation of portfolio returns?

A. 14.91 percent
B. 18.56 percent
C. 21.10 percent
QUESTION SET

9. Lena Hunziger has designed the following three-asset portfolio:

Hunziger estimated the portfolio return to be 6.3 percent. What is the portfolio
standard deviation?
A. 13.07 percent
B. 13.88 percent
C. 14.62 percent
FORECASTING CORRELATION OF RETURNS:
COVARIANCE GIVEN A JOINT PROBABILITY
FUNCTION
u The joint probability function of two random variables X and Y, denoted
P(X,Y), gives the probability of joint occurrences of values of X and Y.
u A formula for computing the covariance between random variables RA and RB
is
𝐶𝑜𝑣 𝑅) , 𝑅* = 0 0 𝑃(𝑅),% , 𝑅*,% ) 𝑅),% − 𝐸𝑅) 𝑅*,& − 𝐸𝑅*
% &
Exhibit 6: Joint Probability Function of
BankCorp and NewBank Returns (Entries Are
Joint Probabilities)
Exhibit 7: Covariance Calculations
FORECASTING CORRELATION OF RETURNS:
COVARIANCE GIVEN A JOINT PROBABILITY
FUNCTION
u Two random variables X and Y are independent if and only if P(X,Y) =
P(X)P(Y).
u Independence is a stronger property than uncorrelatedness because
correlation addresses only linear relationships.
EXAMPLE 2 Covariances and Correlations
of Security Returns
u Isabel Vasquez is reviewing the correlations between four of the asset classes
in her company portfolio. In Exhibit 8, she plots 24 recent monthly returns for
large-cap US stocks versus for large-cap world ex-US stocks (Panel 1) and the
24 monthly returns for intermediate-term corporate bonds versus long-term
corporate bonds (Panel 2). Vasquez presents the returns, variances, and
covariances in decimal form instead of percentage form. Note the different
ranges of their vertical axes (Return %).
Exhibit 8: Monthly Returns for Four Asset
Classes
Exhibit 9: Selected Data for Four Asset
Classes
PORTFOLIO RISK MEASURES: APPLICATIONS OF
THE NORMAL DISTRIBUTION

u In economic theory, mean–variance analysis holds exactly when investors are

risk averse; when they choose investments to maximize expected utility or
satisfaction; and when either (assumption 1) returns are normally distributed
or (assumption 2) investors have quadratic utility functions.
u To illustrate this concept, assume an investor is saving for retirement.
Assuming a long-term expected inflation rate of 2 percent, the minimum
acceptable return would be 2 percent.
Exhibit 10: Probability of Earning a Minimum
Acceptable Return
Panel A: Alternative Portfolio Characteristics
Exhibit 10: Probability of Earning a Minimum
Acceptable Return
Panel B: Likelihoods of Attainting Minimal
Acceptable Return
Exhibit 10: Probability of Earning a Minimum
Acceptable Return
Panel B: Likelihoods of Attainting Minimal
Acceptable Return
PORTFOLIO RISK MEASURES: APPLICATIONS OF
THE NORMAL DISTRIBUTION

u Safety-first rules focus on shortfall risk, the risk that portfolio value (or
portfolio return) will fall below some minimum acceptable level over some
time horizon.
u If returns are normally distributed, the safety-first optimal portfolio
maximizes the safety-first ratio (SFRatio), as follows:
𝑆𝐹𝑅𝑎𝑡𝑖𝑜 = 𝐸 𝑅, − 𝑅- /𝜎,
u For a portfolio with a given safety-first ratio, the probability that its return
will be less than RL is Normal(–SFRatio), and the safety-first optimal portfolio
has the lowest such probability.
EXAMPLE 3 The Safety-First Optimal
Portfolio for a Client
u You are researching asset allocations for a client in Canada with a CAD800,000
portfolio. Although her investment objective is long-term growth, at the end
of a year, she may want to liquidate CAD30,000 of the portfolio to fund
educational expenses. If that need arises, she would like to be able to take
out the CAD30,000 without invading the initial capital of CAD800,000. Exhibit
11 shows three alternative allocations.
Exhibit 11: Mean and Standard Deviation
for Three Allocations (in Percent)
EXAMPLE 3 The Safety-First Optimal
Portfolio for a Client
u Address these questions (assume normality for Questions 2 and 3):
1. Given the client’s desire not to invade the CAD800,000 principal, what is the
shortfall level, RL? Use this shortfall level to answer question 2.
2. According to the safety-first criterion, which of the three allocations is the
best?
A. 0.787037 = (25 − 3.75)/27
B. 0.90625 = (11 − 3.75)/8
C. 0.5125 = (14 − 3.75)/20
3. What is the probability that the return on the safety-first optimal portfolio will
be less than the shortfall level?
EXAMPLE 3 The Safety-First Optimal
Portfolio for a Client
QUESTION SET

u A client has a portfolio of common stocks and fixed-income instruments with a

current value of GBP1,350,000. She intends to liquidate GBP50,000 from the
portfolio at the end of the year to purchase a partnership share in a business.
Furthermore, the client would like to be able to withdraw the GBP50,000
without reducing the initial capital of GBP1,350,000. The following table
shows four alternative asset allocations.
QUESTION SET

u Address the following questions (assume normality for Parts B and C):
A. Given the client’s desire not to invade the GBP1,350,000 principal, what is the
shortfall level, RL? Use this shortfall level to answer Question 2.
B. According to the safety-first criterion, which of the allocations is the best?
C. What is the probability that the return on the safety-first optimal portfolio
will be less than the shortfall level, RL?
QUESTION SET

u A client holding a GBP2,000,000 portfolio wants to withdraw GBP90,000 in one

year without invading the principal. According to Roy’s safety-first criterion,
which of the following portfolio allocations is optimal?