506 Midterm 1 Study Guide

Table of Contents
Logistics..................................................................................................................................1
Concepts.................................................................................................................................2
Key Equations.........................................................................................................................3
Week 2: Optimization......................................................................................................................3
Week 3: Review of Statistics & Distribution Theory..........................................................................3
Week 4: Application in Finance........................................................................................................3
Week 5: Sampling & Sampling Distributions.....................................................................................5
Practice Questions...................................................................................................................8
Week 2: Unconstrained Optimization...............................................................................................8
Week 4: Finance Applications...........................................................................................................8
Week 5: Sampling & Sampling Distributions.....................................................................................8
Practice Questions – Answer Key...........................................................................................10
Week 4: Finance Applications.........................................................................................................10
Week 5: Sampling & Sampling Distributions...................................................................................10

Logistics
 Exam date: Tuesday, October 1st (normal class time)
 Exam format: written, free response calculations – no laptops
 Materials allowed:
o Calculator
o Cheat sheet (front and back, EQUATIONS ONLY – NO WORDS)

Note: this study guide is not comprehensive. Any material from the
lectures, PowerPoint slides, or Brightspace from before Midterm 1 is
eligible to be on the test. We are not liable for any missing information.
Concepts
Week 1: Preliminary Concepts
 Key concepts:
o Types of Data, Data Sources, Data Conversion.
o Basic Calculus of Finance.

Week 2: Optimization
 Key concepts:
o Unconstrained Optimization
 Maximizing total revenue, minimizing total costs, risk-free & risky
asset weights
o Constrained Optimization
o Finance Interpretation of Lagrange Multiplier
o Application to Markowitz Portfolio Theory & Diversification
 How you can practice:
o Unconstrained optimization video & textbook problems under TA Resources
o Constrained optimization textbook review
o Practice calculating optimal weights of 2 random stocks

Week 3: Review of Statistics and Distribution Theory

 Key concepts:
o Continuous Distributions: Normal, t, F, Chi-squared &
Exponential
o Discrete Distributions: Binomial and Poisson
o Central Limit Theorem

Week 4: Application in Finance

 Key concepts:
o Measuring Value at Risk (VaR) for assets or portfolios:
Static VaR, Dynamic VaR, Scaling of VaR, Equity VaR.
 SCALING RULE: If X ~ N(µ, σ 2), then hX ~ N(hµ, h∗σ 2)
o Measuring Downside Risk, Lower Partial Moments (LPM).
o Computing the global minimum variance portfolio (GMVP)
 How you can practice:
o a short practice on VaR (Brightspace)
o Assignment 5

Week 5: Sampling and Sampling Distributions:

 Key concepts:
o Point Estimation
o Properties of Point Estimators; Bias, Efficiency and
Consistency.
 Good estimator: unbiased and efficient
 o Interval Estimation of a Population Mean and Variance.
 Margin of error & interval estimation calculation
o Hypothesis Testing

Key Equations
Week 2: Optimization
Elasticity

 Price elasticity: (absolute value of ep > or = 1  ELASTIC)

% ∆ Q dQ P
ε p= = ×
% ∆ P dP Q
 Total revenue:
TR=P × Q

 Marginal revenue:

d (TR)
M R=T R' =
d (Q)
 Monopoly rules – profit maximization when:
MR=MC

Week 3: Review of Statistics & Distribution Theory

Central Limit Theorem

 Z score
X−µ X−µ
Z= =
SE σ
√n
 Standard error: standard deviation of sample mean
σ
S E=σ X =
√n

Week 4: Application in Finance

Finance Applications 1: Risk-Free Asset v. Risky Asset & Capital Allocation Line
If an investor chooses to to allocate her money between a risk-free assest with a return of r f and a
risky asset with a return r a , then:

 Expected return to the portfolio of the two assets is:

E(r p)=(1−w)r f + w r a

 The risk (variance) to the portfolio of the two asset is:

2 2 2
σ p=w ∗σ a
 Solving from variance equation, the weight of risky asset is:
σp
w=
σa
 Weight of risk-free asset is:
w f =1−w

 Re-writing expected return equation and substituting for w:

E(r a )−r f
E(r p)=r f + σp
σa

 From this, we get the SHARPE RATIO:

(r a−r f )
σp
σa

 From this, we get the formula for CAPITAL ALLOCATION LINE:

E(r p)=r f + λσ

Value at Risk (VaR)

 VaR formula:
VaR=R+(z × s)
 This is a loss (in present value terms) we are fairly
sure will not be exceeded if the current portfolio is
held static over a specified period of time.
 2 parameters: significance level (α ¿ and time
horizon

 Static VaR:
o Assumes no rebalancing over risk horizon  weights are not constant
 Dynamic VaR:
o Assumes portfolio is continually rebalanced  weights are constant

 Equity VaR:
o Risk adjusted portfolio (RAP) is the sum of weighted values of assets in the
portfolio, weighted by corresponding assets’ betas ( β )
 RAP=∑ β i × Ri
o Ri is the return of the asset
EVaR=RAP ×VaR

Converting Return to Z for Calculating VaR

x−x
z=
s

Or

x−µ
z=
SE

** NOTE: use when asked about probability of return  if given VaR %, no need **

Week 5: Sampling & Sampling Distributions

Measures of Central Tendency: Mean, Median, Mode

 Sample mean (point estimate of population mean):

n

∑X
X = i =1
n

Measures of Variation

 Sample variance (point estimate of population variance):

n
s2=∑ ¿ ¿ ¿
i =1
 Sample standard deviation:
s= √ S2
 Range:
Range=X max −X min
 Mean absolute deviation (MAD):
n
MAD=∑ ¿ X i− X∨ ¿ ¿
i=1 n
 Standard error:
 s
S E=s X =
√n
 Population variance:
N
σ =∑ ¿ ¿ ¿
2

i=1

Downside Risk:

 Semi-Variance:
SV =E ¿
 Semi-Standard Deviation:
SSD=√ SV
 Lower Partial Moment:
LPM 2, τ =E ¿
o Here, LPM is the expected value of the minimum squared
deviation from the target return
o Rτ = targeted or expected return

Hypothesis Testing:

 2-sample T test – T statistic formula (unequal variances):

X 1−X 2
t=

√
2 2
S1 S2
+
n1 n2

 2-sample T test – T statistic formula (equal variances):

X 1−X 2
t=

√ 1 1
s 2p ( + )
n 1 n2

 Pooled variance (for equal variance T test):

2 ( n1−1 ) s 21+ ( n2−1 ) s22
s p=
n1 +n2−2
 Degrees of freedom (for equal variance T test):
DF =n1 +n2−2

 F-test statistic
2
s1
F= 2
s2
Interval Estimation:

 Interval estimation:
P ( x−ME ≤ µ ≤ x+ ME )=1−α

 Margin of error:
M E=z α /2 × SE

** HINT: z α/ 2 will almost always be 1.96 (alpha = 0.05, 95% CI) **

Expectations:

 Expected value: same as population mean

n
E( X )=∑ Pi∗X i
i=1

 Expected variance:
2
σ =E ¿
Or:
2
σ =E ¿
Practice Questions

Week 2: Unconstrained Optimization

 See video on Brightspace under TA Resources for textbook questions

Week 4: Finance Applications

 Also see “Short Practice on VaR” on Brightspace under Content

Finance Applications 1: Risk-Free Asset v. Risky Asset & Capital Allocation Line

An investor has allocated 30% of her investment fund on US government bond with an average
annual return of 1.5% and 70% on an ETF with an expected annual return of 6%. The variance of
the ETF is estimated to be .0035. What is the expected return and the expected variance of
the investor’s portfolio? Calculate the Sharpe ratio and write the equation of the CAL. What
would be the expected return to the portfolio if the investor was willing to take 8% risk?

Finance Applications 2: VaR & EVaR

$1M is invested in stock A with a β of 1.2. $2M is invested in stock B with a β of .8. If excess
return on the portfolio is iid and normally distributed with expected mean .05 and expected
SD .20 per year, what is the 1%, 10-day VaR?

Week 5: Sampling & Sampling Distributions

 Also review assignment 4 for measures of central tendency and variation

a. What are the null and alternate hypothesis for testing a hypothesis that the mean
population return of a portfolio is equal to 3% (in percentage terms, e.g. 3% = 3)?
b. What type of test are we conducting in A?
c. Interpret these results. Do we accept or fail to reject the null?
d. What are the null and alternate hypothesis for testing a hypothesis that the
variances of two portfolios are equal?
e. What type of test are we conducting in D?
f. Interpret these results:

g. What are the null and alternate hypothesis for testing a hypothesis that the mean
population returns of two portfolios are equal?
h. What type of test are we conducting in G?
i. Interpret these results:
Practice Questions – Answer Key

Week 4: Finance Applications

Finance Applications 1: Risk-Free Asset v. Risky Asset & Capital Allocation Line

The expected return to new portfilo is E(R p)=(1−w)∗R f + w∗R ETF =.30∗.015+.70∗.06=.0465
. The variance of the portfolio is σ 2p=w 2∗σ 2ETF =.49∗.0035=.001715 . The standard deviation of
.0465−.015
the portfolio is √ .001715=.04141. The Sharpe ratio is, λ= =.76 . Thefore, the
.04141
equation of the CAL can be written as, E(r p)=.015+.76 σ p . If the investor is willing to tolerate
8% risk then, the investor’s expected return is E(r p)=.015+.76∗.08=.0758 or 7.58 .

Finance Application 2: VaR & EVaR

X− X
VaR follows the standard normal distribution Z = . The loss or VaR here will be
SD
X =X + Z∗SD.
1. The 10-day expected excess returns, μ10 = .05*10/250 = .2%.
2. The 10-day SD of the excess returns, σ10 = .2(10/250)^.5 = 4%
3. The 1% 10-day equity VaR: EVaR10,.01 = 2.8(-2.326*4% + .2%) = -254951, where Z =
qnorm(.01, 0, 1, lower.tail=TRUE) = -2.326.

Week 5: Sampling & Sampling Distributions

a. H0: µ = 3
Ha: µ ≠ 3

b. One sample t-test

c. Since the p-value of the test 0.03435 < .05, we fail to accept the null hypothesis.

d. H0: σ 21=σ 22
Ha: σ 21 ≠ σ 22

e. F-test for variances

f. Since the p-value of the test, p-value = 0.5656, we fail to reject the null hypothesis.
g. H0: µx = µy
 Ha: µx ≠ µy
h. Two sample t-test for means
i. Since the p-value of the test, p-value = 0.8153, we fail to reject the null hypothesis.