Quantitative Methods in Finance
137071407
FIN617-chapter11BFQ2020.pptxQuantitative Methods in Finance
FIN 617
Professor Edwalds
Chapter 11 Topic 4
Review Question
What are the assumptions of Arbitrage Pricing Theory?
Answer: There are three assumptions underlying Arbitrage Pricing Theory
Model describes asset returns based on sensitivity to risk factors priced in market
Many assets: diversified portfolios eliminate asset-specific risk
No arbitrage opportunities
Carhart-Fama-French Model
Where
return on the portfolio
risk-free rate
return in excess of expected return based on risk factors (alpha)
return on value-weighted equity index over one-month T-Bills
avg return on small cap over large cap stocks (small minus big)
avg return on high book-to-market stocks over low book-to-market
avg return on top 30% of stocks over bottom 30% (momentum)
sensitivity of portfolio to factor
Carhart-Fama-French Model as APT Model
Multifactor extension of CAPM
Accounts for size, value, and momentum
Risk factors priced in market
Added momentum factor to Fama-French 3-factor model
Multifactor Models
Returns on assets are driven by small number of systematic factors representing priced risk
Explain asset returns better than market model
Types of multifactor models
Macroeconomic factors
Fundamental factors (e.g., P/E, P/B, market cap, leverage)
Statistical factors
Out of scope for this course
Macroeconomic Multifactor Models
An Example
Example 3: Returns Given Factor Sensitivities
Assume inflation & GDP growth explain returns
Returns of two stocks (Manumatic and Nextech) are modeled as
Portfolio weights are 1/3 in M, 2/3 in N
Then
Fundamental Factor Models
Examples of Factors Used
Company Fundamentals
Earnings growth
Earnings variability
Earnings momentum
Financial leverage
Macroeconomic Factors
Industry
Yield curve sensitivity
CAPM beta
Company Share-related Factors
Earnings yield
Dividend yield
Market capitalization
Book-to-market ratio
Share price volatility
Share price momentum
Trading volume
Macroeconomic vs Fundamental
Macroeconomic Factor Models
Model based on difference between actual and expected outcome of factor (surprise)
Expected outcomes already reflected in market prices
Data series of surprises and actual returns
Sensitivities are estimated by regression
Fundamental Factor Models
Factor sensitivities are attributes of securities
Expressed as # of std devs difference from overall mean
Standardized betas
Except for dummy variables
Factor returns are estimated by regression
Homework
Workbook Chapter 11 problem #2
Review Questions
What is the Carhart-Fama-French model?
What do the factors in a macroeconomic multifactor model measure?
What are standardized betas in fundamental multifactor models?
model
factor
by the
explained
not
i
asset
return to
the
of
portion
the
of
portion
the
represents
mean that
zero
with
error term
an
k
factor
in
surprise
a
to
i
asset
on
return
the
of
y
sensitivit
the
b
K
.,
.
.
2,
1,
k
k,
factor
in
surprise
the
F
i
asset
return to
expected
the
a
i
asset
on
return
the
R
F
b
...
F
b
F
b
a
R
i
ik
k
i
i
i
K
iK
2
2
i
1
1
i
i
i
=
e
=
=
=
=
=
e
+
+
+
+
+
=
model
factor
by the
explained
not
i
asset
return to
the
of
portion
the
of
portion
the
represents
mean that
zero
with
error term
an
surprises
growth
GDP
to
i
stock
return to
the
of
y
sensitivit
b
growth
GDP
in
surprise
the
F
surprises
rate
inflation
to
i
stock
return to
the
of
y
sensitivit
b
rate
inflation
in the
surprise
the
F
i
stock
return to
expected
the
a
i
stock
on
return
the
i
i2
GDP
i1
INFL
i
2
1
=
=
=
=
=
=
=
+
+
+
=
e
e
the
the
R
F
b
F
b
a
R
i
i
GDP
i
INFL
i
i
i
model
factor
by the
explained
not
i
asset
return to
the
of
portion
the
of
portion
the
represents
mean that
zero
with
error term
an
i
asset
for
k
attribute
for
average
above
devs
std
of
number
the
b
K
.,
.
.
2,
1,
k
k,
factor
for
dev
std
per
return
extra
the
F
i
asset
for
intercept
regression
the
a
i
asset
on
return
the
...
i
ik
k
i
2
2
1
1
=
=
=
=
=
=
+
+
+
+
+
=
e
e
i
i
K
iK
i
i
i
i
R
F
b
F
b
F
b
a
R
