Portfolio Analysis

Topic 13
I. Markowitz Mean-Variance
Analysis
A. Mean (Expost) vs.
Expected (Exante)

 1. Mean (Expost) Returns are

statistically derived from historical
observations.
 2. Expected (Exante) Returns are
statistically derived expected values
from future estimates of observations.
B. Expected Return of a
Portfolio
 1. The expected return of a portfolio is
a weighted average of the expected
returns of its component securities,
using relative market values as weights.
 Portfolio Analysis

Topic 13
II. Diversification and MPT
A. The Dominance Principle
 States that among all investments with
a given return, the one with the least
risk is desirable; or given the same level
of risk, the one with the highest return is
most desirable.
Dominance Principle Example
 Security E(Ri) 
ATW 7% 3%
GAC 7% 4%
YTC 15% 15%
FTR 3% 3%
HTC 8% 12%
 ATW dominates GAC
 ATW dominates FTR
B. Diversification
 1. Normal Diversification
• This occurs when the investor combines
more than one (1) asset in a portfolio
Risk

Unsystematic 75% of Co.

Risk Total Risk

Systematic Risk 25% of Co.

Total Risk
1 5 10 20 30 # of Assets
Risk
 Unsystematic Risk
• ... is that portion of an asset’s total risk
which can be eliminated through
diversification
 Systematic Risk
• ... is that risk which cannot be eliminated
• Inherent in the marketplace
Diversification
 Superfluous or Naive Diversification
• Occurs when the investor diversifies in
more than 20-30 assets. Diversification for
diversification’s sake.
• a. Results in difficulty in managing such a
large portfolio
• b. Increased costs
– Search and transaction
3. Markowitz Diversification
 This type of diversification considers the
correlation between individual securities.
It is the combination of assets in a portfolio
that are less then perfectly positively
correlated.
• a. The two asset case:
– Stk. A Stk. B
– E(R) 5% 15%
–  10% 20%
 3. Markowitz Diversification
(continued)
 Assume that the investor invests 50% of
capital stock in stock A and 50% in B
• 1. Calculate E(R)
• E(Rp)n = xi E(Ri)
i=1
• E(Rp) =.5(.05) + .5(.15)
• E(Rp) = .025 + .075
• E(Rp) = .10 or 10%
3. Markowitz Diversification
(continued)
 2. Graphically

E(Rp)
B
15%

10% Portfolio AB

5% A

5 10 15 20 25 
3. Markowitz Diversification
(continued)
 Portfolio Return of AB will always be on
line AB depending on the relative fractions
invested in assets A and B.
 3. Calculating the risk of the portfolio
• Consider 3 possible relationships between A
and B
– Perfect Positive Correlation
– Zero Correlation
– Perfect Negative Correlation
Perfect Positive Correlation
 A and B returns vary in identical pattern.
Hence, there is a linear risk-return
relationship between the two assets.
Perfect Positive Correlation
(continued)

E(Rp)
B
15%

AB

5% A

10 15 20 p
Perfect Positive Correlation
(continued)
 Therefore, the risk of portfolio AB is simply
the weighted value of the two assets’ .
• In this case:

p = xA2 A2 + xB2 B2 + 2 xAxBABAB

p = .25(.10)2+.25(.20)2+2(.5)(.5)(.10)(.20)

p = .15 or 15%
Zero Correlation
 A’s return is completely unrelated to B’s
return. With zero correlation, a
substantial amount of risk reduction can
be obtained through diversification.
Zero Correlation (continued)
E(Rp)
B
15%

10% AB

5%
A

10 11.2 20 p
p = .25(.10)2+.25(.20)2

  11.2%
Negative Correlation
 A’s and B’s returns vary perfectly
inversely. The portfolio variance is
always at the lowest risk level
regardless of proportions in each asset.
Negative Correlation (continued)
E(Rp)
B
15%
AB
10%

5%
A

5 10 20 p
p = .25(.10)+.25(.20)+2(.5)(.10)(.20)(-1)

p = .05 or 5%
Markowitz Diversification
 Although there are no securities with
perfectly negative correlation, almost all
assets are less than perfectly
correlated. Therefore, you can reduce
total risk (p) through diversification. If
we consider many assets at various
weights, we can generate the efficient
frontier.
Efficient Frontier Graph
E(Rp)
Efficient
Frontier
M

p
Efficient Frontier
 The Efficient Frontier represents all the
dominant portfolios in risk/return space.
 There is one portfolio (M) which can be
considered the market portfolio if we
analyze all assets in the market. Hence,
M would be a portfolio made up of assets
that correspond to the real relative
weights of each asset in the market.
Efficient Frontier (continued)
 Assume you have 20 assets. With the
help of the computer, you can calculate
all possible portfolio combinations. The
Efficient Frontier will consist of those
portfolios with the highest return given
the same level of risk or minimum risk
given the same return (Dominance
Rule)
Efficient Frontier (continued)
 4. Borrowing and lending investment
funds at R to expand the Efficient
Frontier.
• a. We keep part of our funds in a saving
account
– Lending, OR
• b. We can borrow funds for a greater
investment in the market portfolio
Efficient Frontier (continued)
CML
E(Rp) Borrowing
B
Efficient
Frontier
Lending
M

RF A

p
Portfolio A: 80% of funds in RF, 20% of funds in M
Portfolio B: 80% of funds borrowed to buy more of M,
100% or own funds to buy M
Efficient Frontier (continued)
 By using RF, the Efficient Frontier is now
dominated by the capital market line
(CML). Each portfolio on the capital
market line dominates all portfolios on
the Efficient Frontier at every point
except M.
5. The Portfolio Investment
CML
E(Rp)
Efficient
Frontier
M

Mutual Fund Portfolios

RF with a cash position

p

Investors’ indifference curves are based on their degree

of risk aversion and investment objectives and goals.