FINA 3080:

Investment Analysis and Portfolio Management

Lecture 3 –
Portfolio Mathematics and Capital Allocation

Chao Ying
 Announcements
• Form homework teams of 4-6 students
• Across my two sessions (A and B) is ok
• It is ok if you have a smaller team. We will combine smaller teams
It is also ok if you don’t have a team. We will assign you into one
• Email our TA, not me (!) on or before 11:59pm Friday Sep 20
• Let us know by email if you have any exam conflict
• Our exams are on Oct 17 (7-9pm) & Nov 28 (7-9pm)
• We assume you can take our exams if we don’t hear by Sep 17
• We will arrange the makeup exam for you (including midterm).
• If you can not take the midterm makeup exam, the final exam will
count 80% (35%+45%).

• Quiz Solution & Practice Problems on Blackboard (for your own

practice) 2
 Recap from Last Lecture
• Financial Markets and The Economy

• How Securities Are Traded

• Buying on Margin

• Short Selling

3
Highlights of the debate: https://www.youtube.com/watch?v=2ec4Oa2L1fo

Who win the debate? Let us look at the stock price.

4
↓ sxpaveainves.to 对 harrìs 信
心
 Outline of Today’s Lecture
• Value at Risk (VaR)
• Risky, Riskfree Prospects & The Risk Premium
• Risk Preferences & Returns Utility
• Mean-Variance & Utility Indifference

5
 Another Risk Measure: Value at Risk
最 可能损失

Value at Risk (VaR): Lowest 5% of Returns • VaR is the 5th

percentile of
Under Normal Distribution the distribution
VaR = – 1.65 • There is a 5%
16% 68% 16%
probability
e.g., = 1%, = 2% that the return
VaR = 1% – 1.65 2% will be less than
VaR = 2.3% VaR
5% 90% 5%

retnrndecieaseby-zilotheprobis5%Darelyhappensle.tn
2.5% 95% 2.5%

VaR
0.5% 99% 0.5%
rigmtn
-3 -2 -1 1 2 3 returns 6
大
Realized Returns: r N(R(r),s )
Value at Risk
• If realized returns are rA: arithmetic mean of returns
approximately normally s: standard deviation of returns
distributed,
VaR rA – 1.65s
↓

toreplacehistoricalthe xpectedmeanandriskhis.tn
-3s -2s
VaR

-1s
rA – 1.65 s

R(r) 1s 2s 3s returns
alretum
7
 Non-Normal Returns
Value at Risk
• However, realized • The VaR is
returns often exhibit informative in that it
makes no assumption

Normalsomanyretumsin.hu
“fat tails”, indicating
non-normality on the “true”
distribution
• Mean and
variance • It is still the 5th
no longer fully percentile of
describe returns the distribution

VaR = Lowest 5% of
Returns tail

8
N on_no ma 1 时 要 rank

VaR Example
• You have 100 observations of returns. To calculate the
VaR, the first thing to do is the rank the returns from
highest to lowest:

1 +10.1%
2 +9.8%
…
95 -7% 95% of the time is better than the VaR
96 -7.2% 5% of the time is worse than the VaR
97 -8%
98 … The cutoff point for the VaR can
99 … be either -7%, -7.2% or the
100 … average between
9
 Outline of Today’s Lecture
Value at Risk (VaR)
• Risky, Riskfree Prospects & The Risk Premium
• Risk Preferences & Returns Utility
• Mean-Variance & Utility Indifference

10
 Risky & Riskfree Prospects
Investment State Probability Outcome Return Risk
W s p(s) o(s) r(s)
$150 50%
↑
g ood.de60%
$100 E[r] = 22% 34%
p 40%

$80 -20%
Riskfree Prospect
$105 5% 0%
(Government debt)
priceiscons.tn 11
Simple Prospects & Risk Premium

E[r]
Risky Prospect 22% 34%
–
Riskfree Prospect 5% 0%
=

Risk Premium 17% E[r] - RF

Return (Reward) for Risk Taken

12
 Outline of Today’s Lecture
Value at Risk (VaR)
Risky, Riskfree Prospects & The Risk Premium
• Risk Preferences & Returns Utility
• Mean-Variance & Utility Indifference

13
Risk Preferences & Returns Utility
• Utility Score (or Certainty Equivalent, CE):
• Combines risk aversion, expected returns & risk
• Puts a preference score on securities & portfolios

• Helps to predict how an investor will invest

• Helps advise investors on what they should do

• Formula: U = E[r] – ½ A 2 = CE
Where A measures investor risk aversion
howmucnyondonotiker.sk 14
Utility Scores, Certainty Equivalent
• The (hypothetical) riskfree return that would make
you indifferent between investing in a riskfree
prospect and a risky one.
U = E[r] – ½ A 2= Risk free rate
Ifyouonlynoldriskfraasset.youwihonYF.cn
• Hence: “Certainty Equivalent”: Different people
have different risk aversions

15
whichchoicema.es yonnappīeī

Utility Scores & Investment Choice

• Suppose E[r] = 22%, = 34%, and RF = 5%
• What is the utility score for the risky prospect?
U = 22% – ½ A 34%2, depends on A:

If A = 1, U = 16% > RF Risky prospect

If A = 2, U = 10% > RF Risky prospect
If A = 3, U = 4% < RF Riskfree prospect
Moral: Different investors, different choices
16
 Outline of Today’s Lecture
Value at Risk (VaR)
Risky, Riskfree Prospects & The Risk Premium
Risk Preferences & Returns Utility
• Mean-Variance & Utility Indifference

17
Utility Indifference: Risk Averse
Utility Indifference: The return-risk combinations
that are equally agreeable (i.e., generate the same
utility level) to a given investor.
8.0%
E(r)
6.0%
E[r] = U + ½A 2

4.0%
The risk-averse
2.0%
investor
A U E[r ] demands more
0.0% 0% 4 2% 2.0% return as risk
5% 4 2% 2.5% increases.
-2.0% 10% 4 2% 4.0%
15% 4 2% 6.5%
-4.0%
0.0% 2.5% 5.0% 7.5% 10.0% 12.5% 15.0% 18
Utility Indifference: Risk Neutral
riskdoe.SNtatfact
8.0%
E(r)
The risk-neutral investor
6.0%
demands neither more nor
less return as risk increases.
4.0%
E[r] = U + ½A 2

2.0%
A U E[r ]
0.0%
0% 0 2% 2.0%
5% 0 2% 2.0%
-2.0% 10% 0 2% 2.0%
15% 0 2% 2.0%
-4.0%
0.0% 2.5% 5.0% 7.5% 10.0% 12.5% 15.0% 19
Utility Indifference: Risk Loving
8.0%
E(r) The risk-loving investor
demands less return as A U E[r ]
6.0%
risk increases! 0% -4 2% 2.0%
4.0%
5% -4 2% 1.5%
E[r] = U + ½A 2 10% -4 2% 0.0%
2.0% 15% -4 2% -2.5%

0.0%

-2.0%

-4.0%
0.0% 2.5% 5.0% 7.5% 10.0% 12.5% 15.0%
20
 Utility Indifference Curve

As umingnormaldistributionsamek.ve/hapines Ford1
8.0%
E(r) E[r] = U + ½A 2
6.0%

same
4.0% utility

2.0%
same
std dev

0.0%

肶吐 Risk Averse, A = 4
-2.0%
Risk Neutral, A = 0
ìnvestors Risk Loving, A = -4
-4.0%
0.0% 2.5% 5.0% 7.5% 10.0% 12.5% 15.0%
21
 Mean-Variance Criterion
Return North: Better Portfolios

loniycaresretum7ldublerewards.ir
E(r) if Risk Neutral
Northwest: Better Northeast: Better
Portfolios if Portfolios if
Risk Averse Risk Loving 7

E(rP)

Southeast: Worse
Southwest: Worse Portfolios if Risk Averse
Portfolios
if Risk Loving South: Worse Portfolios
if Risk Neutral Risk
P
22
 Lotteries?
• How do we explain lotteries?
Imagine two situations:
1. Mark Six: Pay HK$10, Receive HK$100m
1
for a small
chance positivere.tn m (
川 州 的)
2. Reverse Mark Six: Receive HK$10, Pay HK$100m for
a small chance

• Game 1 vs. Game 2

• Same standard deviation 不同的 assnmptìons
• Which one you pick: Do not satisfy normal distribution
23
十
比