The Capital Asset Pricing Model
The Capital Asset Pricing Model
¤ The
simplest
and
most
widely
used
model
for
this
computa-on
¨ The
market
porBolio
¨ Beta
risk
E[rE ] − rF
β=
E[rM ] − rF
E[ΔS] E[S t+1 − S t ] E[S t+1 ] − S t
E[rE ] = = =
St St St
CAPM
Equa-ons
from
Topic
4
4
E[S t +1 ]
St =
1 + kE
E[S t +1 ]
St =
1 +
rF +
β ⋅ (Ε[rM ] − rF )
CAPM
Expected
Returns
5
~N(ri,
σi)
¨ Expected
returns
on
the
market
porBolio:
~N(rM,
σM)
¨ Expected
returns
on
a
porBolio:
~N(rP,
σP)
Market
PorBolio
6
rP
Increasing M
σP
The
Market
PorBolio
7
M
rM
rF
Finally
all
investable
risky
assets
are
included
and
the
op&mal
fron-er
becomes
the
efficient
fron-er
σM
The
market
porBolio
is
the
op-mal
risky
asset
when
all
investable
risky
assets
are
included.
The
market
porBolio
weights
are
defined
by
the
rela-ve
total
equity
values
(market
capitaliza-ons)
of
the
risky
assets.
Market
PorBolio
Proxies
8
¨ The
op-mal
risky
porBolio
on
the
efficient
fron-er
given
a
risky
free
asset
is
the
market
por2olio,
M
¤ The
sum
total
of
what
all
investors
own
must
be
op-mal
¤ Investors
are
ra-onal,
have
all
available
informa-on,
have
the
same
investment
horizon
and
sta-s-cal
return
es-mates,
use
mean-‐
variance
op-miza-on,
¤ More
will
be
said
on
‘ra-onal’
and
‘informa-on’
in
topic
12
¨ It
is
ra-onal
for
investors
to
hold
the
market
por2olio
as
their
risky
asset
along
with
the
risk
free
asset
¤ 54%
VEU
¤ 46%
VTI
CML
ST
r
SM
CAL
M
T
rF
σ
Capital
Market
Line
13
CML
Market
PorBolio
rM
ri
Si
CAL
SM
Si
=
(ri
–
rF)
/
σi
rF
Sharpe
ra-o
SM
=
(rM
–
rF)
/
σM
for
asset
i
Sharpe
ra-o
for
market
porBolio
σM
σi
Deriva-on
of
the
Beta
Risk
Factor
14
M M
σ εij ≡ σ ij − βiβ jσ M2
σ
P2 = ∑ ⋅ (β ⋅ β ⋅ σ 2
∑ i j i j M +σεij )
w w
i=1 j=1
M M M M
2 2
σ =∑ P
i=1
∑w w β β σ + ∑ ∑w w σ
j=1
i j i j M
i=1 j=1
i j εij
Deriva-on
of
the
Beta
Factor
15
M M M M
¨ Split
σP2 = ∑
∑
w
j
β
i
β
jσM2
+ ∑ ∑
i
w
wiw jσ
εij
⎛⎜ M M M
⎞ ⎛ M
⎟ ⎜ M M
⎞
⎟
σ
P2 = ⎜ ∑ w2iβ2iσM2 + ∑ ∑ wiw jβiβ jσM2 ⎟ + ⎜ ∑ w2iσ2ε + ∑ ∑ wiw jσε i ij ⎟
⎜ i=1 i=1 j=1 ⎟ ⎜ i=1 i=1 j=1 ⎟
⎝ j≠ i ⎠ ⎝ j≠ i ⎠
variance
covariance
variance
covariance
¨ Firm
specific
covariance
is
assumed
zero.
Split
the
M M M
2 2 2 2 2 2
σ = ∑ w (β σ + σ ) + ∑
P i i M εi w w
∑ i jijM
β β σ
i=1 i=1 j=1
j≠ i
2 2 2 2
σij = βiβ jσM2
Systemic
σ =β σ +σ
i i M εi
risk
only
2
Systemic
and
σiM = β β σ i M M
non-‐systemic
2
(firm
specific)
σiM = β σ i M
risk
σiM
βi = 2
σM
Deriva-on
of
the
Beta
Factor
17
σiM
βi = 2
σM
Sub
into
CAPM
formula
Price
of
risk
ri − rF rM − rF
σiM = 2
ri = rF + 2 (rM − rF ) σiM σM
σM
σiM = ρiMσi σM
σi
βi = ρiM
σM
PorBolio
Beta
Deriva-on
18
σ PM
βP = 2
σM
cov(rP ,rM )
βP = 2
σM
M
cov(∑ w iri ,rM )
βP = i=1
2
σM
M M
∑ w cov(r ,r
i i M ) βP = ∑ wi ⋅ βi
βP = i=1
2
i=1
σ M
M
∑ (w σ i iM )
βP = i=1
2
σM
Security
Characteris8c
Line
19
15.0%
Plot
of
Walmart
vs.
SPX
excess
10.0% returns
from
Jan
2001
to
Nov
2005
(r
–
rF)
β=.637
5.0% α=.0003
SCL
β
0.0%
-12.5% -10.0% -7.5% -5.0% -2.5% 0.0% 2.5% 5.0% 7.5% 10.0%
Ordinary
Least
-5.0% Squares
Assump-ons
εk
-10.0% εk ~ N(0, σ 2ε )
cor(ε k , ε k +1 ) = 0
-15.0%
σ2i = β2iσM2 + σ2εi
(rM
–
rF)
(r − rF ) = β(rM − rF ) + α + εk
Security
Characteris-c
Line
20
(r – rF )
(r – rF ) = β(rM – rF ) + α + εk
εk
β
β·∙(rM
–
rF
)
(rM – rF )
εk ~ N(0, σ 2ε )
E[εk ] = 0
CAPM
Model
21
σ2i = β2iσM
2
+ σ2εi
β2iσM2 β2iσM2
ex − ante : (r − rF ) = β(rM − rF ) R = 2 = 2 2
2
σi βi σM + σ 2εi
ex − post :
(r − rF ) = β(rM − rF ) + α + εk Yahoo
Finance
SML
ex-‐ante
22
r
TM
ri − rF rM − rF
TM = =
rM
βi βM
βM = 1
ri
ri − rF = βi ⋅ (rM − rF )
rF
βi
βM β
SML
ex-‐post
23
r
According
to
CAPM
if
rM
M
αi
<
0,
the
asset
is
under-‐priced
assets
overpriced
and
should
ri
rF+βj(rM-rF) be,
shorted,
or
under-‐
αi weighted
(Ti
<
TM)
-αj
rF+βi(rM-rF) α i
>
0,
the
asset
is
underpriced
and
rj
should
be
bought
or
over-‐weighted
(Ti
>
TM)
over-‐priced
α i
=
0,
the
asset
is
fairly
rF
assets
priced
according
to
CAPM
(Ti
=
TM)
βi
βj
βM
β
CAPM
parameter
interpreta-on
24
¨ A
high
posi-ve
beta
for
an
asset
has
the
following
interpreta-on.
¤ The
asset
will
have
large
price
swings
driven
by
market,
SPX,
movements
¤ The
asset
will
increase
the
risk
in
the
investor’s
porBolio
¤ The
investor
will
expect
a
high
return
¤ The
asset
will
outperform
in
a
rising
market
¨ Various
studies
show
that
posi-ve
alpha
is
open
associated
with
¤ Low
β
stocks
EB
¤ High
(value
stocks)
E
¤ Small
cap
stocks
¤ High
dividend
yield
stocks
Implica-ons
of
CAPM
25
¨ The
market
price
of
risk
is
the
same
for
all
properly
priced
securi-es
and
porBolios
¨ Investors
will
choose
to
hold
combina-ons
of
the
market
porBolio
and
the
risk
free
asset
¨ The
market
porBolio
is
on
the
efficient
fron-er
¨ Only
systema-c
risk
is
priced
into
an
asset
¨ For
an
individual
stock,
the
only
risk
that
brings
excess
return
is
the
risk
that
the
stock
contributes
to
the
market
porBolio.
CAPM
Model
Assump-ons
26
Market
Assump8ons
1. All
assets
globally
are
traded
(can
be
shorted)
and
divisible
2. For
every
borrower,
there
is
a
lender
&
supply
=
demand
3. There
is
a
riskless
security
4. No
taxes
and
transac-on
costs
5. Investors
are
price
takers
6. Assets
returns
normally
distributed
(characterized
by
two
parameters)
7. All
investors
borrow
and
lend
at
the
riskless
rate
Investor
assump8ons
1. Ra-onal,
risk
averse
and
maximize
expected
u-lity
of
return
2. U-lity
is
perceived
as
risk
adjusted
return
3. Risk
measured
as
standard
devia-on
of
return
4. Single
period
-me
horizon
5. Homogeneous
sta-s-cal
return
expecta-ons
Essen-al
Concepts
27