Asset Pricing:Linear and Positive Pricing
Law of one price(LOOP) and No Arbitrage : restrictions to Market Price
- LOOP : if
h
,
k
∈
R
J
,
X
?
h
=
X
?
k
→
P
?
h
=
P
?
k
h,k\in R^J,X·h=X·k\to P·h=P·k
h,k∈RJ,X?h=X?k→P?h=P?k??(具有相同收益的所有投资组合价格相同)
- NSA : if
X
h
≥
0
,
t
h
e
n
?
p
h
≥
0
Xh\geq0,then \ ph\geq0
Xh≥0,then?ph≥0?
- NA : if
X
h
>
0
,
t
h
e
n
?
p
h
>
0
Xh>0,then\ ph>0
Xh>0,then?ph>0?
如果没有冗余证券,只有一个投资组合可以生成任意给定收益,因此一价定律自然满足。如果存在冗余证券,一价定律可能被满足也可能不被满足,取决于证券价格。
Lemma 1 : LOOP
→
\to
→?? if
X
h
=
0
,
t
h
e
n
?
p
h
=
0
Xh=0,then\ ph=0
Xh=0,then?ph=0???
proof:
X
h
=
X
k
→
X
(
h
?
k
)
p
h
=
p
k
→
p
(
h
?
k
)
h
?
k
=
w
→
p
w
=
0
,
x
w
=
0
Xh=Xk\to X(h-k)\\ph=pk\to p(h-k)\\h-k=w\to pw=0,xw=0
Xh=Xk→X(h?k)ph=pk→p(h?k)h?k=w→pw=0,xw=0
Lemma 2 : NA
→
\to
→?? NSA(显然)
Lemma 3 : NSA
→
\to
→? LOOP
proof:考虑 Contrapositive Argument : NSA
→
\to
→? LOOP
?
\Leftrightarrow
?? LOOP!
→
\to
→?? NSA!
If LOOP!, then
?
h
,
k
∈
R
J
,
X
h
=
X
k
,
p
h
≠
p
k
\exist h,k\in R^J,Xh=Xk,ph\neq pk
?h,k∈RJ,Xh=Xk,ph?=pk?
-
p
h
>
p
k
→
p
(
k
?
h
)
<
0
,
x
(
k
?
h
)
=
0
,
k
?
h
=
w
→
N
S
A
!
ph>pk\to p(k-h)<0,x(k-h)=0,k-h=w\to NSA!
ph>pk→p(k?h)<0,x(k?h)=0,k?h=w→NSA!?
-
p
h
<
p
k
→
p
(
h
?
k
)
<
0
,
x
(
h
?
k
)
=
0
,
h
?
k
=
q
→
N
S
A
!
ph<pk\to p(h-k)<0,x(h-k)=0,h-k=q\to NSA!
ph<pk→p(h?k)<0,x(h?k)=0,h?k=q→NSA!??
Lemma 4 : suppose LOOP ! ,
?
?
z
∈
M
,
z
=
X
h
,
p
h
\exist \ z\in M,z=Xh,ph
??z∈M,z=Xh,ph can be anything.
proof:Already show : LOOP
→
\to
→? 0 payoff portfolio has 0 price.
If LOOP!,零收益组合可以按任意价格购买。
Payoff Pricing Functional(收益定价泛函):
Given
p
∈
R
J
p\in R^J
p∈RJ?,定义映射
q
:
M
→
R
q:M\to R
q:M→R:
q
(
z
)
≡
{
w
:
w
=
p
h
,
z
=
h
X
,
f
o
r
?
s
o
m
e
?
h
∈
R
J
}
q(z)\equiv \{w:w=ph,z=hX,for\ some\ h\in R^J\}
q(z)≡{w:w=ph,z=hX,for?some?h∈RJ}
定理2.3.1:
q
q
q??? is linear functional iff LOOP holds.
proof:if LOOP holds
→
\to
→ q(0)=0 , q(z) is single value.
suppose h,h’,考虑收益
z
=
h
X
,
z
′
=
h
′
X
z=hX,z'=h'X
z=hX,z′=h′X。对任意
α
,
β
∈
R
\alpha,\beta\in R
α,β∈R,
α
z
+
β
z
′
=
α
h
X
+
β
h
′
X
\alpha z+\beta z'=\alpha hX+\beta h'X
αz+βz′=αhX+βh′X,
q
(
α
z
+
β
z
′
)
=
p
(
α
h
+
β
h
′
)
=
α
p
h
+
β
p
h
′
=
α
q
(
z
)
+
β
q
(
z
′
)
q(\alpha z+\beta z')=p(\alpha h+\beta h')=\alpha ph+\beta ph'=\alpha q(z)+\beta q(z')
q(αz+βz′)=p(αh+βh′)=αph+βph′=αq(z)+βq(z′)?,所以
q
q
q 是线性的。
若一价定律成立,称
q
q
q 为收益定价泛函。三个算子:
(
q
(
z
)
,
p
,
X
)
(q(z),p,X)
(q(z),p,X),每个组合
h
h
h 是所有证券持有量的
J
J
J 维向量,所有组合的集合
R
J
R^J
RJ?? 称为投资组合空间(portfolio space)。证券价格向量
p
p
p 可以看成组合空间
R
J
R^J
RJ 到实数的线性泛函:
p
:
R
J
→
R
p:R^J\to R
p:RJ→R 类似地,收益矩阵
X
X
X 看成从组合空间
R
J
R^J
RJ 到资产张成空间
M
M
M 的线性算子:
X
:
R
J
→
M
?
R
S
X:R^J\to M\subseteq R^S
X:RJ→M?RS 对每个组合
h
h
h 指定收益
h
X
hX
hX,
q
q
q 是泛函,则有:
p
=
q
?
°
?
X
?
h
∈
R
J
,
h
p
=
q
(
h
X
)
p=q\ \circ\ X\\\forall h\in R^J,hp=q(hX)
p=q?°?X?h∈RJ,hp=q(hX) 线性均衡定价(Linear Equilibrium Pricing):
定理2.4.1:If
u
(
c
0
,
c
1
)
u(c_0,c_1)
u(c0?,c1?) is strictly increasing in period 0, then in Equilibrium , LOOP hold,
q
(
z
)
q(z)
q(z) is linear.
proof:prof : Suppose LOOP fails ,
?
h
0
,
h
0
X
=
0
,
p
h
0
≠
0
\exist h_0,h_0X=0,ph_0\neq0
?h0?,h0?X=0,ph0??=0??. if
p
h
0
<
0
,
?
(
h
,
c
0
,
c
1
)
?
f
e
a
s
i
b
l
e
ph_0<0,\forall(h,c_0,c_1)\ feasible
ph0?<0,?(h,c0?,c1?)?feasible,
u
(
c
0
?
p
h
0
,
c
1
)
>
u
(
c
0
,
c
1
)
,
(
h
+
h
0
,
c
0
?
p
h
0
,
c
1
)
?
f
e
a
s
i
b
l
e
.
u(c_0-ph_0,c_1)>u(c_0,c_1),(h+h_0,c_0-ph_0,c_1)\ feasible.
u(c0??ph0?,c1?)>u(c0?,c1?),(h+h0?,c0??ph0?,c1?)?feasible.预算可行且严格偏好。
如果0期消费不进入个体的效用函数,Market Equilibrium 2:
max
?
(
c
1
i
,
h
i
)
u
i
(
c
1
)
p
?
h
≤
w
0
c
1
≤
w
1
+
h
?
X
∑
i
=
1
N
h
i
=
0
\max_{(c_1^i,h^i)} u^i(c_1)\\p·h\leq w_0\\c_1\leq w_1+h·X\\\sum_{i=1}^Nh_i=0
(c1i?,hi)max?ui(c1?)p?h≤w0?c1?≤w1?+h?Xi=1∑N?hi?=0
定理2.4.2 : if u is strictly increasing in period 1,and
?
h
∈
R
J
,
s
.
t
?
h
?
X
>
0
\exist h\in R^J,s.t\ h·X>0
?h∈RJ,s.t?h?X>0, then in Equilibrium , LOOP holds ,
q
(
z
)
q(z)
q(z)? is linear.
proof:suppose LOOP fails ,
?
h
′
,
s
.
t
?
h
′
X
=
0
,
p
h
′
≠
0.
h
0
?
X
>
0
,
?
α
∈
R
,
α
p
?
h
′
=
p
?
h
0
\exist h',s.t\ h'X=0,ph'\neq0.h_0·X>0,\exist\alpha\in R,\alpha p·h'=p·h_0
?h′,s.t?h′X=0,ph′?=0.h0??X>0,?α∈R,αp?h′=p?h0?
?
h
,
(
h
+
h
0
?
α
h
′
)
?
h
\forall h,(h+h_0-\alpha h')\succ h
?h,(h+h0??αh′)?h(组合
(
h
+
h
0
?
α
h
′
)
(h+h_0-\alpha h')
(h+h0??αh′) 与组合
h
h
h 花费相同,但生产出的消费束严格偏好于
h
h
h 所生产的。所以最优组合不存在,矛盾。)
Example :
J
=
3
,
x
1
=
(
1
,
0
)
,
x
2
=
(
0
,
1
)
,
x
3
=
(
1
,
1
)
,
s
=
2
J=3,x_1=(1,0),x_2=(0,1),x_3=(1,1),s=2
J=3,x1?=(1,0),x2?=(0,1),x3?=(1,1),s=2
u
(
c
0
,
c
11
,
c
12
)
=
?
(
c
0
?
1
)
2
?
(
c
11
?
1
)
2
?
(
c
12
?
2
)
2
,
w
0
=
1
,
w
1
=
(
1
,
2
)
u(c_0,c_{11},c_{12})=-(c_0-1)^2-(c_{11}-1)^2-(c_{12}-2)^2,w_0=1,w_1=(1,2)
u(c0?,c11?,c12?)=?(c0??1)2?(c11??1)2?(c12??2)2,w0?=1,w1?=(1,2)
x
1
+
x
2
=
x
3
→
[
1
,
1
,
0
]
[
x
1
x
2
x
3
]
=
[
0
,
0
,
1
]
[
x
1
x
2
x
3
]
x_1+x_2=x_3\to\left[\begin{matrix}1,1,0\end{matrix}\right]\left[\begin{matrix}x_1\\x_2\\x_3\end{matrix}\right]=\left[\begin{matrix}0,0,1\end{matrix}\right]\left[\begin{matrix}x_1\\x_2\\x_3\end{matrix}\right]
x1?+x2?=x3?→[1,1,0?]???x1?x2?x3?????=[0,0,1?]???x1?x2?x3??????
If LOOP holds,
[
1
,
1
,
0
]
[
x
1
x
2
x
3
]
=
[
0
,
0
,
1
]
[
x
1
x
2
x
3
]
→
[
p
1
,
p
2
,
p
3
]
[
1
1
0
]
=
[
p
1
,
p
2
,
p
3
]
[
0
0
1
]
→
p
1
+
p
2
=
p
3
\left[\begin{matrix}1,1,0\end{matrix}\right]\left[\begin{matrix}x_1\\x_2\\x_3\end{matrix}\right]=\left[\begin{matrix}0,0,1\end{matrix}\right]\left[\begin{matrix}x_1\\x_2\\x_3\end{matrix}\right]\to\left[\begin{matrix}p_1,p_2,p_3\end{matrix}\right]\left[\begin{matrix}1\\1\\0\end{matrix}\right]=\left[\begin{matrix}p_1,p_2,p_3\end{matrix}\right]\left[\begin{matrix}0\\0\\1\end{matrix}\right]\to p_1+p_2=p_3
[1,1,0?]???x1?x2?x3?????=[0,0,1?]???x1?x2?x3?????→[p1?,p2?,p3??]???110????=[p1?,p2?,p3??]???001????→p1?+p2?=p3?
因为禀赋是饱和点,任意证券价格
p
1
,
p
2
,
p
3
p_1,p_2,p_3
p1?,p2?,p3? 都是均衡价格。
完备市场状态价格(State Price in Complete Market)
s种状态:令
e
s
e_s
es? 表示未定权益的空间
R
S
R^S
RS 的第
s
s
s 个基向量,称为状态
s
s
s 的状态权益或阿罗证券(Arrow securities):
M
=
R
S
,
e
1
=
[
1
0
?
0
]
,
e
2
=
[
0
1
?
0
]
,
?
?
,
e
S
=
[
0
0
?
1
]
M=R^S,e_1=\left[\begin{matrix}1\\0\\\vdots\\0\end{matrix}\right],e_2=\left[\begin{matrix}0\\1\\\vdots\\0\end{matrix}\right],\cdots,e_S=\left[\begin{matrix}0\\0\\\vdots\\1\end{matrix}\right]
M=RS,e1?=??????10?0???????,e2?=??????01?0???????,?,eS?=??????00?1??????? If
M
=
R
S
M=R^S
M=RS(市场完备), LOOP holds,定义:
q
s
≡
q
(
e
s
)
q_s\equiv q(e_s)
qs?≡q(es?) 为状态
s
s
s 的state price。
e
s
∈
M
=
R
S
e_s\in M=R^S
es?∈M=RS
?
z
∈
M
=
R
S
,
q
(
z
)
=
q
?
z
,
q
=
[
q
1
q
2
?
q
S
]
\forall z\in M=R^S,q(z)=q·z,q=\left[\begin{matrix}q_1\\q_2\\\vdots\\q_S\end{matrix}\right]
?z∈M=RS,q(z)=q?z,q=??????q1?q2??qS????????
对于每个证券的价格:
p
j
=
q
?
x
j
→
p
=
X
?
q
→
q
=
L
p
p_j=q·x_j\to p=X·q\to q=Lp
pj?=q?xj?→p=X?q→q=Lp?
q
(
z
)
=
q
?
z
=
p
?
h
,
h
=
[
0
0
?
1
0
]
j
+
1
,
p
?
h
=
p
j
=
q
(
z
)
=
q
?
x
j
q(z)=q·z=p·h,h=\left[\begin{matrix}0\\0\\\vdots\\1\\0\end{matrix}\right]_{j+1},p·h=p_j=q(z)=q·x_j
q(z)=q?z=p?h,h=????????00?10?????????j+1?,p?h=pj?=q(z)=q?xj?
|