文章目录
知识补充
导数
原函数 | 导函数 |
---|---|
y = ln x y = \ln x y=lnx | y ′ = 1 x y' = \frac{1}{x} y′=x1 |
y = x n y = x^n y=xn | y ′ = n x n − 1 y' = nx^{n-1} y′=nxn−1 |
y = 1 x y = \frac{1}{x} y=x1 | y ′ = − 1 x 2 y' = -\frac{1}{x^2} y′=−x21 |
y = f [ g ( x ) ] y = f[g(x)] y=f[g(x)] | y ′ = f ′ [ g ( x ) ] ⋅ g ′ ( x ) y' = f'[g(x)] \cdot g'(x) y′=f′[g(x)]⋅g′(x) |
凹凸性
一元函数
设
f
(
x
)
f(x)
f(x) 在区间
I
I
I 上连续,如果对
I
I
I 上任意两点
x
1
x_1
x1、
x
2
x_2
x2,恒有
f
(
x
1
+
x
2
2
)
<
f
(
x
1
)
+
f
(
x
2
)
2
,
f \left(\frac{x_1+x_2}{2} \right) < \frac{f(x_1)+f(x_2)}{2},
f(2x1+x2)<2f(x1)+f(x2),
那么称
f
(
x
)
f(x)
f(x) 在
I
I
I 上的图形是凹的(或凹弧);
如果恒有
f
(
x
1
+
x
2
2
)
>
f
(
x
1
)
+
f
(
x
2
)
2
,
f \left(\frac{x_1+x_2}{2} \right) > \frac{f(x_1)+f(x_2)}{2},
f(2x1+x2)>2f(x1)+f(x2),
那么称
f
(
x
)
f(x)
f(x) 在
I
I
I 上的图形是凸的(或凸弧)。
二元函数
设
f
(
x
,
y
)
f(x, y)
f(x,y) 在区域
D
D
D 上具有二阶连续偏导数,记
A
=
f
x
x
′
′
(
x
,
y
)
A=f''_{xx}(x,y)
A=fxx′′(x,y),
B
=
f
x
y
′
′
(
x
,
y
)
B=f''_{xy}(x,y)
B=fxy′′(x,y),
C
=
f
y
y
′
′
(
x
,
y
)
C=f''_{yy}(x,y)
C=fyy′′(x,y),则:
(1)在
D
D
D 上恒有
A
>
0
A>0
A>0,且
A
C
−
B
2
≥
0
AC-B^2 \geq 0
AC−B2≥0 时,
f
(
x
,
y
)
f(x,y)
f(x,y) 在区域
D
D
D 上是凸函数;
(2)在
D
D
D 上恒有
A
<
0
A<0
A<0,且
A
C
−
B
2
≥
0
AC-B^2 \geq 0
AC−B2≥0 时,
f
(
x
,
y
)
f(x,y)
f(x,y) 在区域
D
D
D 上是凹函数。
设 f ( x , y ) f(x,y) f(x,y) 是在开区域 D D D 内具有连续偏导数的凸(或者凹)函数, ( x 0 , y 0 ) ∈ D (x_0,y_0)\in D (x0,y0)∈D 且 f x ′ ( x 0 , y 0 ) = 0 f'_x(x_0,y_0)=0 fx′(x0,y0)=0, f y ′ ( x 0 , y 0 ) = 0 f'_y(x_0,y_0)=0 fy′(x0,y0)=0,则 f ( x 0 , y 0 ) f(x_0,y_0) f(x0,y0) 必为 f ( x , y ) f(x,y) f(x,y) 在 D D D 内的最小值(或最大值)。
三大经典生产函数
- 柯布道格拉斯生产函数
- 莱昂提夫生产函数
- CES生产函数
柯布道格拉斯函数
公式:
Y
=
A
L
β
K
α
,
Y = AL^{\beta}K^{\alpha},
Y=ALβKα,
其中:
符号 | 意义 |
---|---|
Y Y Y | 产出数量 |
L L L | 劳动力投入 |
K K K | 资本投入 |
A A A | 总生产率因子 |
α \alpha α | 资本的产出弹性 |
β \beta β | 劳动力产出弹性 |
α \alpha α 和 β \beta β 具有以下的关系:
关系 | 意义 |
---|---|
α + β = 1 \alpha + \beta = 1 α+β=1 | 规模报酬不变 |
α + β > 1 \alpha + \beta > 1 α+β>1 | 规模报酬递增 |
α + β < 1 \alpha + \beta < 1 α+β<1 | 规模报酬递减 |
莱昂提夫生产函数
公式:
q
=
min
(
z
1
a
,
z
2
b
)
,
q = \min \left(\frac{z_1}{a}, \frac{z_2}{b} \right),
q=min(az1,bz2),
其中:
符号 | 意义 |
---|---|
z 1 z_1 z1、 z 2 z_2 z2 | 两种要素的投入数量 |
a a a、 b b b | 代表生产技术的系数 |
CES 生产函数
公式:
y
=
[
a
1
x
1
ρ
+
a
2
x
2
ρ
]
1
ρ
.
y = \left[a_1x_1^{\rho} + a_2x_2^{\rho} \right]^{\frac{1}{\rho}}.
y=[a1x1ρ+a2x2ρ]ρ1.
替代弹性:
σ
=
d
ln
x
2
/
x
1
d
ln
∣
T
R
S
∣
=
1
1
−
ρ
.
\sigma = \frac{d\ln x_2 / x_1}{d\ln \left|TRS \right|} = \frac{1}{1 - \rho}.
σ=dln∣TRS∣dlnx2/x1=1−ρ1.
生产技术的模型表达
边际技术替代率
在维持产量水平不变的条件下,增加一个单位的某种要素投入量时所减少的另一种要素的投入量,被称为边际技术替代率(Marginal Rate of Technical Substitution,MRTS)。
边际替代率的计算公式
边际技术替代率 = 等产量曲线的斜率的绝对值 = 保持产量不变的情况下,减少
x
2
x_2
x2 投入同时增加
x
1
x_1
x1 投入数量的比值。
M
R
T
S
(
x
1
,
x
2
)
=
−
Δ
x
2
Δ
x
1
=
M
P
1
M
P
2
.
MRTS(x_1, x_2) = - \frac{\Delta x_2}{\Delta x_1} = \frac{MP_1}{MP_2}.
MRTS(x1,x2)=−Δx1Δx2=MP2MP1.
其中,
M
P
1
MP_1
MP1 和
M
P
2
MP_2
MP2 分别是投入要素 1 和 2 的边际产出。
边际技术替代率的特点
从几何意义上看,在一条等产量曲线上的任意一点,投入
x
1
x_1
x1 对投入
x
2
x_2
x2 的边际技术替代率,等于等产量曲线上这一点的斜率。
边际技术替代率具有如下特点:
- 当等产量曲线的斜率为负值时,表明两种生产要素可以互相替代,一种生产要素增加,另外一种生产要素必须减少方能使产量维持在同一水平上。
- 当等产量曲线的斜率为正值时,表明两种生产要素必须同时增加才能达到与从前相同的产量水平。
- 等产量曲线的斜率也可以是无穷大或为零,此时表明两种生产要素不能互相替代。
替代弹性
替代弹性 = 等产量曲线的曲度 = 要素投入比例相对于边际技术替代率的敏感度。
σ
21
=
d
ln
(
x
2
/
x
1
)
d
ln
M
R
T
S
12
=
d
ln
(
x
2
/
x
1
)
d
ln
(
d
f
d
x
1
/
d
f
d
x
2
)
.
\sigma_{21} = \frac{d\ln(x_2 / x_1)}{d\ln MRTS_{12}} = \frac{d\ln(x_2 / x_1)}{d\ln \left(\frac{df}{dx_1} / \frac{df}{dx_2} \right)}.
σ21=dlnMRTS12dln(x2/x1)=dln(dx1df/dx2df)dln(x2/x1).
注
意
(
我
的
好
友
提
醒
我
)
{\color{blue}注意(我的好友提醒我)}
注意(我的好友提醒我):
例子
考虑道格拉斯生产函数
f
(
x
1
,
x
2
)
=
x
1
a
x
2
1
−
α
f(x_1, x_2) = x_1^{a}x_2^{1-\alpha}
f(x1,x2)=x1ax21−α。
边际技术替代率是:
M
R
T
S
21
=
1
−
a
a
x
1
x
2
MRTS_{21} = \frac{1- a}{a} \frac{x_1}{x_2}
MRTS21=a1−ax2x1
求解过程
{\color{red}\text{求解过程}}
求解过程:
M
R
T
S
(
x
1
,
x
2
)
=
M
P
1
M
P
2
=
∂
f
∂
x
1
/
∂
f
∂
x
2
=
(
x
2
1
−
a
⋅
a
⋅
x
1
a
−
1
)
/
(
x
1
a
⋅
(
1
−
a
)
⋅
x
2
−
a
)
=
a
1
−
a
⋅
x
1
a
−
1
x
1
a
⋅
x
2
1
−
a
x
2
−
a
=
a
1
−
a
⋅
1
x
1
⋅
x
2
1
=
a
1
−
a
x
2
x
1
.
\begin{aligned} MRTS(x_1, x_2) &= \frac{MP_1}{MP_2} \\ &= \frac{\partial f}{\partial x_1} / \frac{\partial f}{\partial x_2} \\ &= \left(x_2^{1-a}\cdot a \cdot x_1^{a-1} \right) / \left(x_1^a \cdot (1-a) \cdot x_2^{-a} \right) \\ &= \frac{a}{1-a} \cdot \frac{x_1^{a-1}}{x_1^a} \cdot \frac{x_2^{1-a}}{x_2^{-a}} \\ &= \frac{a}{1-a} \cdot \frac{1}{x_1} \cdot \frac{x_2}{1} \\ &= \frac{a}{1-a} \frac{x_2}{x_1} \end{aligned}.
MRTS(x1,x2)=MP2MP1=∂x1∂f/∂x2∂f=(x21−a⋅a⋅x1a−1)/(x1a⋅(1−a)⋅x2−a)=1−aa⋅x1ax1a−1⋅x2−ax21−a=1−aa⋅x11⋅1x2=1−aax1x2.
M
R
T
S
(
x
1
,
x
2
)
MRTS(x_1, x_2)
MRTS(x1,x2) 简记为
M
R
T
S
12
MRTS_{12}
MRTS12,
M
R
T
S
21
=
M
P
2
M
P
1
=
1
M
R
T
S
12
=
1
−
a
a
x
1
x
2
.
MRTS_{21} = \frac{MP_2}{MP_1}=\frac{1}{MRTS_{12}}=\frac{1-a}{a}\frac{x_1}{x_2}.
MRTS21=MP1MP2=MRTS121=a1−ax2x1.
记符号
θ
\theta
θ 为
θ
=
1
−
a
a
x
1
x
2
,
\theta = \frac{1- a}{a} \frac{x_1}{x_2},
θ=a1−ax2x1,
则有
x
1
x
2
=
a
1
−
a
θ
.
\frac{x_1}{x_2} = \frac{a}{1-a} \theta.
x2x1=1−aaθ.
替代弹性为
σ
21
=
d
ln
(
x
1
x
2
)
d
ln
(
M
R
T
S
21
)
=
d
ln
(
x
1
x
2
)
d
ln
(
θ
)
=
d
x
1
x
2
x
1
x
2
θ
d
θ
=
d
x
1
x
2
d
θ
θ
x
1
x
2
=
a
1
−
a
1
−
a
a
x
1
x
2
x
2
x
1
=
1.
\begin{aligned} \sigma_{21} &= \frac{d\ln \left(\frac{x_1}{x_2} \right)}{d\ln \left(MRTS_{21}\right)} \\ &= \frac{d\ln \left(\frac{x_1}{x_2} \right)}{d\ln \left(\theta \right)} \\ &= \frac{d\frac{x_1}{x_2}}{\frac{x_1}{x_2}}\frac{\theta}{d\theta} \\ &= \frac{d\frac{x_1}{x_2}}{d\theta}\frac{\theta}{\frac{x_1}{x_2}} \\ &= \frac{a}{1-a}\frac{1-a}{a}\frac{x_1}{x_2}\frac{x_2}{x_1} \\ &= 1. \end{aligned}
σ21=dln(MRTS21)dln(x2x1)=dln(θ)dln(x2x1)=x2x1dx2x1dθθ=dθdx2x1x2x1θ=1−aaa1−ax2x1x1x2=1.
求
解
方
法
二
{\color{red}求解方法二}
求解方法二:
σ
21
=
d
ln
(
x
1
x
2
)
d
ln
(
M
R
T
S
21
)
=
d
ln
(
a
1
−
a
θ
)
d
ln
θ
=
(
d
ln
(
a
1
−
a
θ
)
d
θ
)
/
(
d
ln
θ
d
θ
)
=
(
1
a
1
−
a
θ
⋅
a
1
−
a
)
/
(
1
θ
)
=
(
1
θ
)
/
(
1
θ
)
=
1.
\begin{aligned} \sigma_{21} &= \frac{d\ln \left(\frac{x_1}{x_2} \right)}{d\ln \left(MRTS_{21}\right)} \\ &= \frac{d\ln (\frac{a}{1-a} \theta)}{d \ln \theta} \\ &= \left(\frac{d \ln (\frac{a}{1-a} \theta)}{d\theta}\right) / \left(\frac{d\ln \theta}{d \theta}\right) \\ &= \left(\frac{1}{\frac{a}{1-a} \theta} \cdot \frac{a}{1-a}\right) / \left(\frac{1}{\theta} \right)\\ &= \left(\frac{1}{\theta}\right) / \left(\frac{1}{\theta}\right) \\ &= 1. \end{aligned}
σ21=dln(MRTS21)dln(x2x1)=dlnθdln(1−aaθ)=(dθdln(1−aaθ))/(dθdlnθ)=(1−aaθ1⋅1−aa)/(θ1)=(θ1)/(θ1)=1.
CES生产函数的替代弹性?
求
解
过
程
{\color{red}求解过程}
求解过程:
CES生产函数为
y
=
[
a
1
x
1
ρ
+
a
2
x
2
ρ
]
1
ρ
y = \left[a_1x_1^{\rho}+a_2x_2^{\rho} \right]^{\frac{1}{\rho}}
y=[a1x1ρ+a2x2ρ]ρ1
边际技术替代率 MRTS 为
M
R
T
S
12
=
M
P
1
M
P
2
=
(
∂
y
∂
x
1
)
/
(
∂
y
∂
x
2
)
=
(
1
ρ
[
a
1
x
1
ρ
+
a
2
x
2
ρ
]
1
ρ
−
1
)
⋅
a
1
ρ
x
1
ρ
−
1
(
1
ρ
[
a
1
x
1
ρ
+
a
2
x
2
ρ
]
1
ρ
−
1
)
⋅
a
2
ρ
x
2
ρ
−
1
=
a
1
a
2
(
x
1
x
2
)
ρ
−
1
.
\begin{aligned} MRTS_{12} &= \frac{MP_1}{MP_2} \\ &=\left(\frac{\partial y}{\partial x_1} \right) / \left(\frac{\partial y}{\partial x_2} \right) \\ &=\frac{\left(\frac{1}{\rho}\left[a_1x_1^{\rho}+a_2x_2^{\rho}\right]^{\frac{1}{\rho}-1} \right) \cdot a_1\rho x_1^{\rho-1}}{\left(\frac{1}{\rho}\left[a_1x_1^{\rho}+a_2x_2^{\rho} \right]^{\frac{1}{\rho}-1} \right) \cdot a_2 \rho x_2^{\rho-1}} \\ &= \frac{a_1}{a_2}\left(\frac{x_1}{x_2}\right)^{\rho-1}. \end{aligned}
MRTS12=MP2MP1=(∂x1∂y)/(∂x2∂y)=(ρ1[a1x1ρ+a2x2ρ]ρ1−1)⋅a2ρx2ρ−1(ρ1[a1x1ρ+a2x2ρ]ρ1−1)⋅a1ρx1ρ−1=a2a1(x2x1)ρ−1.
记符号
θ
\theta
θ 为
θ
=
a
1
a
2
(
x
1
x
2
)
ρ
−
1
.
\theta = \frac{a_1}{a_2}\left(\frac{x_1}{x_2}\right)^{\rho-1}.
θ=a2a1(x2x1)ρ−1.
则有
x
1
x
2
=
(
a
2
a
1
θ
)
1
ρ
−
1
.
\frac{x_1}{x_2} = \left(\frac{a_2}{a_1}\theta\right)^{\frac{1}{\rho-1}}.
x2x1=(a1a2θ)ρ−11.
替代弹性
σ
12
=
d
ln
(
x
2
x
1
)
d
ln
(
M
R
T
S
12
)
=
d
ln
(
x
2
x
1
)
d
ln
θ
=
d
ln
(
1
(
a
2
a
1
θ
)
1
ρ
−
1
)
d
ln
θ
=
(
d
ln
(
1
(
a
2
a
1
θ
)
1
ρ
−
1
)
d
θ
)
/
(
d
ln
θ
d
θ
)
=
(
a
2
a
1
θ
)
1
ρ
−
1
⋅
(
−
1
)
1
(
(
a
2
a
1
θ
)
1
ρ
−
1
)
2
⋅
1
ρ
−
1
(
a
2
a
1
θ
)
1
ρ
−
1
−
1
⋅
(
a
2
a
1
)
/
(
1
θ
)
=
(
−
1
)
1
ρ
−
1
⋅
(
a
2
a
1
θ
)
−
1
⋅
(
a
2
a
1
)
/
(
1
θ
)
=
(
−
1
)
1
ρ
−
1
θ
−
1
θ
=
(
−
1
)
1
ρ
−
1
=
1
1
−
ρ
.
\begin{aligned} \sigma_{12} &= \frac{d\ln\left(\frac{x_2}{x_1}\right)}{d\ln \left(MRTS_{12} \right)} \\ &= \frac{d\ln\left(\frac{x_2}{x_1}\right)}{d\ln \theta} \\ &= \frac{d\ln \left(\frac{1}{\left(\frac{a_2}{a_1}\theta\right)^{\frac{1}{\rho-1}}} \right)}{d\ln\theta} \\ &= \left(\frac{d\ln \left(\frac{1}{\left(\frac{a_2}{a_1}\theta\right)^{\frac{1}{\rho-1}}} \right)}{d \theta } \right) / \left(\frac{d \ln\theta }{d\theta}\right) \\ &= \left(\frac{a_2}{a_1}\theta \right)^{\frac{1}{\rho-1}} \cdot (-1)\frac{1}{\left(\left(\frac{a_2}{a_1}\theta \right)^{\frac{1}{\rho-1}} \right)^2} \cdot \frac{1}{\rho-1} \left(\frac{a_2}{a_1}\theta \right)^{\frac{1}{\rho-1}-1} \cdot \left(\frac{a_2}{a_1}\right) / \left(\frac{1}{\theta}\right) \\ &= (-1)\frac{1}{\rho-1} \cdot \left(\frac{a_2}{a_1}\theta \right)^{-1} \cdot \left(\frac{a_2}{a_1}\right) / \left(\frac{1}{\theta}\right) \\ &= (-1)\frac{1}{\rho-1} \theta^{-1} \theta \\ &= (-1)\frac{1}{\rho-1} \\ &= \frac{1}{1-\rho}. \end{aligned}
σ12=dln(MRTS12)dln(x1x2)=dlnθdln(x1x2)=dlnθdln((a1a2θ)ρ−111)=⎝⎜⎜⎜⎜⎛dθdln((a1a2θ)ρ−111)⎠⎟⎟⎟⎟⎞/(dθdlnθ)=(a1a2θ)ρ−11⋅(−1)((a1a2θ)ρ−11)21⋅ρ−11(a1a2θ)ρ−11−1⋅(a1a2)/(θ1)=(−1)ρ−11⋅(a1a2θ)−1⋅(a1a2)/(θ1)=(−1)ρ−11θ−1θ=(−1)ρ−11=1−ρ1.
练习:
生产函数
q
(
x
1
,
x
2
)
=
x
1
α
x
2
β
q(x_1, x_2) = x_1^{\alpha}x_2^{\beta}
q(x1,x2)=x1αx2β,
α
>
0
\alpha > 0
α>0,
β
>
0
\beta > 0
β>0,
证明(1)当
α
+
β
≤
1
\alpha + \beta \leq 1
α+β≤1 时,
q
(
x
1
,
x
2
)
q(x_1, x_2)
q(x1,x2) 是凹函数。
(2)当
α
>
0
\alpha > 0
α>0,
β
>
0
\beta > 0
β>0 时,
ln
(
q
(
x
1
,
x
2
)
)
\ln \left(q(x_1, x_2) \right)
ln(q(x1,x2)) 是凹函数。
证
明
{\color{red}证明}
证明:
(1)
q
x
1
′
(
x
1
,
x
2
)
=
x
2
β
⋅
α
x
1
α
−
1
=
α
x
1
α
−
1
x
2
β
,
q'_{x_1}(x_1, x_2) = x_2^{\beta} \cdot \alpha x_1^{\alpha-1}=\alpha x_1^{\alpha-1}x_2^{\beta},
qx1′(x1,x2)=x2β⋅αx1α−1=αx1α−1x2β,
q
x
2
′
(
x
1
,
x
2
)
=
x
1
α
⋅
β
x
2
β
−
1
=
β
x
1
α
x
2
β
−
1
,
q'_{x_2}(x_1, x_2) = x_1^{\alpha} \cdot \beta x_2^{\beta-1}=\beta x_1^{\alpha}x_2^{\beta-1},
qx2′(x1,x2)=x1α⋅βx2β−1=βx1αx2β−1,
A
=
q
x
1
x
1
′
′
(
x
1
,
x
2
)
=
α
x
2
β
⋅
(
α
−
1
)
x
1
α
−
2
=
α
(
α
−
1
)
x
1
α
−
2
x
2
β
,
A = q''_{x_1x_1}(x_1, x_2) = \alpha x_2^{\beta} \cdot (\alpha-1) x_1^{\alpha-2}=\alpha (\alpha-1)x_1^{\alpha-2}x_2^{\beta},
A=qx1x1′′(x1,x2)=αx2β⋅(α−1)x1α−2=α(α−1)x1α−2x2β,
B
=
q
x
1
x
2
′
′
(
x
1
,
x
2
)
=
α
x
1
α
−
1
⋅
β
x
2
β
−
1
=
α
β
x
1
α
−
1
x
2
β
−
1
,
B = q''_{x_1x_2}(x_1,x_2) = \alpha x_1^{\alpha-1} \cdot \beta x_2^{\beta-1} = \alpha \beta x_1^{\alpha-1} x_2^{\beta-1},
B=qx1x2′′(x1,x2)=αx1α−1⋅βx2β−1=αβx1α−1x2β−1,
C
=
q
x
2
x
2
′
′
(
x
1
,
x
2
)
=
β
x
1
α
⋅
(
β
−
1
)
x
2
β
−
2
=
β
(
β
−
1
)
x
1
α
x
2
β
−
2
,
C = q''_{x_2x_2}(x_1,x_2) = \beta x_1^{\alpha} \cdot (\beta-1)x_2^{\beta-2} =\beta (\beta-1)x_1^{\alpha}x_2^{\beta-2},
C=qx2x2′′(x1,x2)=βx1α⋅(β−1)x2β−2=β(β−1)x1αx2β−2,
A
C
−
B
2
=
α
(
α
−
1
)
x
1
α
−
2
x
2
β
⋅
β
(
β
−
1
)
x
1
α
x
2
β
−
2
−
(
α
β
x
1
α
−
1
x
2
β
−
1
)
2
=
α
β
(
α
−
1
)
(
β
−
1
)
x
1
2
α
−
2
x
2
2
β
−
2
−
α
2
β
2
x
1
2
(
α
−
1
)
x
2
2
(
β
−
1
)
=
(
α
2
β
2
−
α
2
β
−
α
β
2
+
α
β
)
x
1
2
α
−
2
x
2
2
β
−
2
−
α
2
β
2
x
1
2
α
−
2
x
2
2
β
−
2
=
(
−
α
2
β
−
α
β
2
+
α
β
)
x
1
2
α
−
2
x
2
2
β
−
2
=
α
β
(
−
α
−
β
+
1
)
x
1
2
α
−
2
x
2
2
β
−
2
,
\begin{aligned} AC-B^2 &=\alpha (\alpha-1)x_1^{\alpha-2}x_2^{\beta} \cdot \beta (\beta-1)x_1^{\alpha}x_2^{\beta-2} - \left(\alpha \beta x_1^{\alpha-1} x_2^{\beta-1} \right)^2 \\ &= \alpha \beta (\alpha-1)(\beta-1)x_1^{2\alpha-2}x_2^{2\beta-2} - \alpha^2\beta^2x_1^{2(\alpha-1)}x_2^{2(\beta-1)} \\ &= (\alpha^2\beta^2-\alpha^2\beta-\alpha\beta^2+\alpha\beta) x_1^{2\alpha-2}x_2^{2\beta-2} - \alpha^2\beta^2x_1^{2\alpha-2}x_2^{2\beta-2} \\ &= (-\alpha^2\beta-\alpha\beta^2+\alpha\beta)x_1^{2\alpha-2}x_2^{2\beta-2} \\ &= \alpha\beta(-\alpha-\beta+1)x_1^{2\alpha-2}x_2^{2\beta-2}, \\ \end{aligned}
AC−B2=α(α−1)x1α−2x2β⋅β(β−1)x1αx2β−2−(αβx1α−1x2β−1)2=αβ(α−1)(β−1)x12α−2x22β−2−α2β2x12(α−1)x22(β−1)=(α2β2−α2β−αβ2+αβ)x12α−2x22β−2−α2β2x12α−2x22β−2=(−α2β−αβ2+αβ)x12α−2x22β−2=αβ(−α−β+1)x12α−2x22β−2,
因为
α
+
β
≤
1
\alpha+\beta \leq 1
α+β≤1,所以
1
−
α
−
β
≥
0
1-\alpha-\beta \geq 0
1−α−β≥0;
α
>
0
\alpha >0
α>0,
β
>
0
\beta > 0
β>0,所以
α
β
>
0
\alpha \beta > 0
αβ>0。
故
A
C
−
B
2
≥
0
AC-B^2 \geq 0
AC−B2≥0。
因为
0
<
α
<
1
0 < \alpha < 1
0<α<1,所以
α
−
1
<
0
\alpha - 1 < 0
α−1<0,所以
A
<
0
A < 0
A<0。
由二元函数的凹凸性判定可知,
q
(
x
1
,
x
2
)
q(x_1,x_2)
q(x1,x2) 是凹函数。
(2)
记
t
(
x
1
,
x
2
)
=
ln
(
q
(
x
1
,
x
2
)
)
=
ln
(
x
1
α
x
2
β
)
t(x_1,x_2)=\ln \left(q(x_1,x_2) \right)=\ln \left(x_1^{\alpha}x_2^{\beta} \right)
t(x1,x2)=ln(q(x1,x2))=ln(x1αx2β)。
t
x
1
′
(
x
1
,
x
2
)
=
1
x
1
α
x
2
β
⋅
x
2
β
α
x
1
α
−
1
=
α
x
1
−
1
,
t'_{x_1}(x_1, x_2)=\frac{1}{x_1^{\alpha}x_2^{\beta}} \cdot x_2^{\beta}\alpha x_1^{\alpha-1}=\alpha x_1^{-1},
tx1′(x1,x2)=x1αx2β1⋅x2βαx1α−1=αx1−1,
t
x
2
′
(
x
1
,
x
2
)
=
1
x
1
α
x
2
β
⋅
x
1
α
β
x
2
β
−
1
=
β
x
2
−
1
,
t'_{x_2}(x_1,x_2)=\frac{1}{x_1^{\alpha}x_2^{\beta}} \cdot x_1^{\alpha}\beta x_2^{\beta-1}=\beta x_2^{-1},
tx2′(x1,x2)=x1αx2β1⋅x1αβx2β−1=βx2−1,
A
=
t
x
1
x
1
′
′
(
x
1
,
x
2
)
=
α
⋅
(
−
1
)
x
1
−
2
=
−
α
x
1
−
2
,
A = t''_{x_1x_1}(x_1,x_2)=\alpha \cdot(-1)x_1^{-2}= -\alpha x_1^{-2},
A=tx1x1′′(x1,x2)=α⋅(−1)x1−2=−αx1−2,
B
=
t
x
1
x
2
′
′
(
x
1
,
x
2
)
=
0
,
B = t''_{x_1x_2}(x_1,x_2)=0,
B=tx1x2′′(x1,x2)=0,
C
=
t
x
2
x
2
′
′
(
x
1
,
x
2
)
=
β
⋅
(
−
1
)
x
2
−
2
=
−
β
x
2
−
2
,
C = t''_{x_2x_2}(x_1,x_2) = \beta \cdot (-1)x_2^{-2} = -\beta x_2^{-2},
C=tx2x2′′(x1,x2)=β⋅(−1)x2−2=−βx2−2,
A
C
−
B
2
=
(
−
α
x
1
−
2
)
(
−
β
x
2
−
2
)
−
0
=
α
β
(
x
1
x
2
)
−
2
≥
0.
AC-B^2=\left(-\alpha x_1^{-2} \right) \left(-\beta x_2^{-2} \right) - 0=\alpha \beta (x_1x_2)^{-2} \geq 0.
AC−B2=(−αx1−2)(−βx2−2)−0=αβ(x1x2)−2≥0.
并且
A
=
−
α
x
1
−
2
<
0
A = -\alpha x_1^{-2} < 0
A=−αx1−2<0,所以由二元函数凹凸性判定定理,得
t
(
x
1
,
x
2
)
=
ln
(
x
1
α
x
2
β
)
t(x_1,x_2)=\ln \left(x_1^{\alpha}x_2^{\beta} \right)
t(x1,x2)=ln(x1αx2β) 是凹函数,也即是
ln
q
(
x
1
,
x
2
)
\ln q(x_1, x_2)
lnq(x1,x2) 是凹函数。
利润最大化
- 完全竞争市场下
- 简化了市场需求,产品售价给定
- 线性要素成本,没有规模效应
- 仅考虑生产技术约束
- 利润最大化条件:每种要素的边际产出价值等于边际成本
例子:单一产品,产量
y
y
y;单一投入,数量
x
x
x。
生产函数:
y
=
x
a
,
y = x^a,
y=xa,
利润最大化条件:
p
a
x
a
−
1
=
w
pax^{a-1}=w
paxa−1=w
投入要素数量:
x
(
p
,
w
)
=
(
w
a
p
)
1
a
−
1
x(p, w)=\left(\frac{w}{ap}\right)^{\frac{1}{a-1}}
x(p,w)=(apw)a−11
生产产品数量:
y
(
p
,
w
)
=
x
a
(
p
,
w
)
=
(
w
a
p
)
a
a
−
1
y(p, w)=x^a(p, w) = \left(\frac{w}{ap} \right)^{\frac{a}{a-1}}
y(p,w)=xa(p,w)=(apw)a−1a
利润函数:
π
(
p
,
w
)
=
p
y
(
p
,
w
)
−
w
x
(
p
,
w
)
=
p
(
w
a
p
)
a
a
−
1
−
w
(
w
a
p
)
1
a
−
1
=
p
(
w
a
p
)
1
a
−
1
(
w
a
p
)
1
−
w
(
w
a
p
)
1
a
−
1
=
(
w
a
−
w
)
(
w
a
p
)
1
a
−
1
=
w
(
1
−
a
a
)
(
w
a
p
)
1
a
−
1
.
\begin{aligned} \pi(p, w) &= py(p,w)-wx(p,w) \\ &= p\left(\frac{w}{ap} \right)^{\frac{a}{a-1}}-w\left(\frac{w}{ap}\right)^{\frac{1}{a-1}} \\ &= p\left(\frac{w}{ap} \right)^{\frac{1}{a-1}} \left(\frac{w}{ap} \right)^1 - w\left(\frac{w}{ap}\right)^{\frac{1}{a-1}} \\ &= \left(\frac{w}{a}-w \right)\left(\frac{w}{ap} \right)^{\frac{1}{a-1}} \\ &= w\left(\frac{1-a}{a} \right) \left(\frac{w}{ap} \right)^{\frac{1}{a-1}}. \end{aligned}
π(p,w)=py(p,w)−wx(p,w)=p(apw)a−1a−w(apw)a−11=p(apw)a−11(apw)1−w(apw)a−11=(aw−w)(apw)a−11=w(a1−a)(apw)a−11.
(我的理解:产品单价为 p p p,数量为 y y y;投入单价为 w w w,数量为 x x x)
成本最小化
- 可进一步应用于非竞争性市场
min x w x s.t. f ( x ) = y . \min_{\mathbf{x}} \mathbf{wx} \\ \textbf{s.t. } f(\mathbf{x})=y. xminwxs.t. f(x)=y.
成本最小化条件:
w
i
w
j
=
∂
f
(
x
∗
)
∂
x
i
∂
f
(
x
∗
)
∂
x
j
\frac{w_i}{w_j} = \frac{\frac{\partial f(\mathbf{x}^\ast)}{\partial x_i}}{\frac{\partial f(\mathbf{x}^\ast)}{\partial x_j}}
wjwi=∂xj∂f(x∗)∂xi∂f(x∗)
成本函数
生产技术的经济特征
-
柯布道格拉斯技术
c ( w , y ) = min x 1 , x 2 w 1 x 1 + w 2 x 2 s.t. A x 1 a x 2 b = y . c(\mathbf{w}, y) = \min_{x_1, x_2} w_1x_1+w_2x_2 \\ \textbf{s.t. } Ax_1^ax_2^b = y. c(w,y)=x1,x2minw1x1+w2x2s.t. Ax1ax2b=y.
从 柯 布 道 格 拉 斯 生 产 函 数 推 导 成 本 函 数 过 程 {\color{red}从柯布道格拉斯生产函数推导成本函数过程} 从柯布道格拉斯生产函数推导成本函数过程:
第一步:从约束条件中得到 x 2 x_2 x2 的表达式
A x 1 a x 2 b = y x 2 b = y A x 1 a x 2 = ( y A x 1 a ) 1 b . Ax_1^ax_2^b = y \\ x_2^b = \frac{y}{Ax_1^a} \\ x_2 = \left(\frac{y}{Ax_1^a} \right)^{\frac{1}{b}}. Ax1ax2b=yx2b=Ax1ayx2=(Ax1ay)b1.
第二步:将 x 2 x_2 x2 代入优化目标中
c ( w 1 , w 2 , y ) = w 1 x 1 + w 2 x 2 = w 1 x 1 + w 2 ( y A x 1 a ) 1 b . \begin{aligned} c(w_1, w_2, y) &= w_1x_1 + w_2x_2 \\ &= w_1x_1 + w_2 \left(\frac{y}{Ax_1^a} \right)^{\frac{1}{b}} \end{aligned}. c(w1,w2,y)=w1x1+w2x2=w1x1+w2(Ax1ay)b1.
第三步:对 x 1 x_1 x1 求导数,
记
t ( x 1 ) = w 1 x 1 + w 2 ( y A x 1 a ) 1 b , t(x_1) = w_1x_1 + w_2 \left(\frac{y}{Ax_1^a} \right)^{\frac{1}{b}}, t(x1)=w1x1+w2(Ax1ay)b1,
t ′ ( x 1 ) = w 1 + w 2 1 b ( y A x 1 a ) 1 b − 1 ⋅ ( y A ) ( − 1 ) ( 1 x 1 a ) 2 ⋅ a x 1 a − 1 = w 1 + w 2 1 b ( y A ) 1 b − 1 ( 1 x 1 a ) 1 b − 1 ( y A ) ( − 1 ) ( 1 x 1 a ) 2 a x 1 a − 1 = w 1 − w 2 a b ( y A ) 1 b − 1 + 1 ( x 1 − a ) 1 b − 1 + 2 x 1 a − 1 = w 1 − a b w 2 ( y A ) 1 b x 1 − a + b b . \begin{aligned} t'(x_1) &= w_1+w_2 \frac{1}{b} \left(\frac{y}{Ax_1^a} \right)^{\frac{1}{b}-1}\cdot \left(\frac{y}{A} \right) (-1)\left(\frac{1}{x_1^a} \right)^2 \cdot ax_1^{a-1} \\ &=w_1 + w_2 \frac{1}{b} \left(\frac{y}{A}\right)^{\frac{1}{b}-1}\left(\frac{1}{x_1^a} \right)^{\frac{1}{b}-1}\left(\frac{y}{A}\right) (-1) \left(\frac{1}{x_1^a} \right)^2ax_1^{a-1} \\ &= w_1-w_2\frac{a}{b} \left(\frac{y}{A} \right)^{\frac{1}{b}-1+1}\left(x_1^{-a} \right)^{\frac{1}{b}-1+2}x_1^{a-1} \\ &= w_1 - \frac{a}{b}w_2\left(\frac{y}{A} \right)^{\frac{1}{b}}x_1^{-\frac{a+b}{b}}. \end{aligned} t′(x1)=w1+w2b1(Ax1ay)b1−1⋅(Ay)(−1)(x1a1)2⋅ax1a−1=w1+w2b1(Ay)b1−1(x1a1)b1−1(Ay)(−1)(x1a1)2ax1a−1=w1−w2ba(Ay)b1−1+1(x1−a)b1−1+2x1a−1=w1−baw2(Ay)b1x1−ba+b.
令 t ′ ( x 1 ) = 0 t'(x_1) = 0 t′(x1)=0,得
w 1 − a b w 2 ( y A ) 1 b x 1 − a + b b = 0 w 1 = a b w 2 ( y A ) 1 b x 1 − a + b b b w 1 = a w 2 ( y A ) 1 b x 1 − a + b b b w 1 a w 2 = ( y A ) 1 b x 1 − a + b b b w 1 a w 2 ( y A ) − 1 b = x 1 − a + b b 1 x 1 a + b b = b w 1 a w 2 ( y A ) − 1 b ( 1 x 1 ) a + b b = b w 1 a w 2 ( y A ) − 1 b ( 1 x 1 ) = [ b w 1 a w 2 ( y A ) − 1 b ] b a + b ( 1 x 1 ) = ( b w 1 a w 2 ) b a + b [ ( y A ) − 1 b ] b a + b ( 1 x 1 ) = ( b w 1 a w 2 ) b a + b ( y A ) − 1 a + b x 1 = 1 ( b w 1 a w 2 ) b a + b ( y A ) − 1 a + b x 1 = ( a w 2 b w 1 ) b a + b ( A y ) − 1 a + b = A − 1 a + b ( a w 2 b w 1 ) b a + b y 1 a + b . w_1 - \frac{a}{b}w_2\left(\frac{y}{A} \right)^{\frac{1}{b}}x_1^{-\frac{a+b}{b}} = 0 \\ w_1 = \frac{a}{b}w_2\left(\frac{y}{A} \right)^{\frac{1}{b}}x_1^{-\frac{a+b}{b}} \\ bw_1 = aw_2\left(\frac{y}{A} \right)^{\frac{1}{b}} x_1^{-\frac{a+b}{b}} \\ \frac{bw_1}{aw_2} = \left(\frac{y}{A} \right)^{\frac{1}{b}} x_1^{-\frac{a+b}{b}} \\ \frac{bw_1}{aw_2}\left(\frac{y}{A} \right)^{-\frac{1}{b}}=x_1^{-\frac{a+b}{b}} \\ \frac{1}{x_1^{\frac{a+b}{b}}}=\frac{bw_1}{aw_2}\left(\frac{y}{A} \right)^{-\frac{1}{b}} \\ \left(\frac{1}{x_1}\right)^{\frac{a+b}{b}}=\frac{bw_1}{aw_2}\left(\frac{y}{A} \right)^{-\frac{1}{b}} \\ \left(\frac{1}{x_1}\right)=\left[\frac{bw_1}{aw_2}\left(\frac{y}{A} \right)^{-\frac{1}{b}}\right]^{\frac{b}{a+b}} \\ \left(\frac{1}{x_1}\right)=\left(\frac{bw_1}{aw_2}\right)^{\frac{b}{a+b}}\left[\left(\frac{y}{A} \right)^{-\frac{1}{b}}\right]^{\frac{b}{a+b}} \\ \left(\frac{1}{x_1}\right)=\left(\frac{bw_1}{aw_2}\right)^{\frac{b}{a+b}} \left(\frac{y}{A} \right)^{-\frac{1}{a+b}} \\ x_1=\frac{1}{\left(\frac{bw_1}{aw_2}\right)^{\frac{b}{a+b}} \left(\frac{y}{A} \right)^{-\frac{1}{a+b}}} \\ x_1 = \left(\frac{aw_2}{bw_1} \right)^{\frac{b}{a+b}} \left(\frac{A}{y} \right)^{-\frac{1}{a+b}} =A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{\frac{b}{a+b}}y^{\frac{1}{a+b}}. w1−baw2(Ay)b1x1−ba+b=0w1=baw2(Ay)b1x1−ba+bbw1=aw2(Ay)b1x1−ba+baw2bw1=(Ay)b1x1−ba+baw2bw1(Ay)−b1=x1−ba+bx1ba+b1=aw2bw1(Ay)−b1(x11)ba+b=aw2bw1(Ay)−b1(x11)=[aw2bw1(Ay)−b1]a+bb(x11)=(aw2bw1)a+bb[(Ay)−b1]a+bb(x11)=(aw2bw1)a+bb(Ay)−a+b1x1=(aw2bw1)a+bb(Ay)−a+b11x1=(bw1aw2)a+bb(yA)−a+b1=A−a+b1(bw1aw2)a+bbya+b1.
因此
x 1 ( w 1 , w 2 , y ) = A − 1 a + b ( a w 2 b w 1 ) b a + b y 1 a + b . x_1(w_1, w_2, y) =A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{\frac{b}{a+b}}y^{\frac{1}{a+b}}. x1(w1,w2,y)=A−a+b1(bw1aw2)a+bbya+b1.
将 x 1 x_1 x1 带回 x 2 x_2 x2 中,得
x 2 = ( y A x 1 a ) 1 b = y 1 b A − 1 b x 1 − a b = y 1 b A − 1 b [ A − 1 a + b ( a w 2 b w 1 ) b a + b y 1 a + b ] − a b = y 1 b A − 1 b A a b ( a + b ) ( a w 2 b w 1 ) − a a + b y − a b ( a + b ) = A − 1 a + b ( a w 2 b w 1 ) − a a + b y 1 a + b . \begin{aligned} x_2 &= \left(\frac{y}{Ax_1^a} \right)^{\frac{1}{b}} \\ &= y^{\frac{1}{b}}A^{-\frac{1}{b}} x_1^{-\frac{a}{b}} \\ &= y^{\frac{1}{b}}A^{-\frac{1}{b}} \left[A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{\frac{b}{a+b}}y^{\frac{1}{a+b}} \right]^{-\frac{a}{b}} \\ &= y^{\frac{1}{b}}A^{-\frac{1}{b}} A^{\frac{a}{b(a+b)}} \left(\frac{aw_2}{bw_1} \right)^{-\frac{a}{a+b}} y^{-\frac{a}{b(a+b)}} \\ &= A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{-\frac{a}{a+b}} y^{\frac{1}{a+b}}. \end{aligned} x2=(Ax1ay)b1=yb1A−b1x1−ba=yb1A−b1[A−a+b1(bw1aw2)a+bbya+b1]−ba=yb1A−b1Ab(a+b)a(bw1aw2)−a+bay−b(a+b)a=A−a+b1(bw1aw2)−a+baya+b1.
因此
x 2 ( w 1 , w 2 , y ) = A − 1 a + b ( a w 2 b w 1 ) − a a + b y 1 a + b . x_2(w_1, w_2, y) = A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{-\frac{a}{a+b}} y^{\frac{1}{a+b}}. x2(w1,w2,y)=A−a+b1(bw1aw2)−a+baya+b1.
第四步:将 x 1 x_1 x1、 x 2 x_2 x2 代入原成本函数中
c ( w 1 , w 2 , y ) = w 1 x 1 + w 2 x 2 = w 1 A − 1 a + b ( a w 2 b w 1 ) b a + b y 1 a + b + w 2 A − 1 a + b ( a w 2 b w 1 ) − a a + b y 1 a + b = A − 1 a + b ( a b ) b a + b w 2 b a + b w 1 a a + b y 1 a + b + A − 1 a + b ( a b ) − a a + b w 2 b a + b w 1 a a + b y 1 a + b = A − 1 a + b w 1 a a + b w 2 b a + b y 1 a + b [ ( a b ) b a + b + ( a b ) − a a + b ] . \begin{aligned} c(w_1, w_2, y) &= w_1x_1+w_2x_2 \\ &= w_1A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{\frac{b}{a+b}}y^{\frac{1}{a+b}} +w_2A^{-\frac{1}{a+b}} \left(\frac{aw_2}{bw_1} \right)^{-\frac{a}{a+b}} y^{\frac{1}{a+b}} \\ &=A^{-\frac{1}{a+b}}\left(\frac{a}{b}\right)^{\frac{b}{a+b}}w_2^{\frac{b}{a+b}}w_1^{\frac{a}{a+b}}y^{\frac{1}{a+b}}+A^{-\frac{1}{a+b}}\left(\frac{a}{b} \right)^{-\frac{a}{a+b}}w_2^{\frac{b}{a+b}}w_1^{\frac{a}{a+b}} y^{\frac{1}{a+b}}\\ &=A^{-\frac{1}{a+b}}w_1^{\frac{a}{a+b}}w_2^{\frac{b}{a+b}}y^{\frac{1}{a+b}} \left[\left(\frac{a}{b}\right)^{\frac{b}{a+b}} +\left(\frac{a}{b}\right)^{-\frac{a}{a+b}}\right]. \\ \end{aligned} c(w1,w2,y)=w1x1+w2x2=w1A−a+b1(bw1aw2)a+bbya+b1+w2A−a+b1(bw1aw2)−a+baya+b1=A−a+b1(ba)a+bbw2a+bbw1a+baya+b1+A−a+b1(ba)−a+baw2a+bbw1a+baya+b1=A−a+b1w1a+baw2a+bbya+b1[(ba)a+bb+(ba)−a+ba].
通常:令 A = 1 A=1 A=1, a + b = 1 a+b=1 a+b=1,则上式简化为:
c ( w 1 , w 2 , y ) = K w 1 a w 2 1 − a y , c(w_1,w_2,y)=Kw_1^aw_2^{1-a}y, c(w1,w2,y)=Kw1aw21−ay,
其中 K = a − a ( 1 − a ) a − 1 K=a^{-a}(1-a)^{a-1} K=a−a(1−a)a−1。 -
莱昂提夫技术
f ( x 1 , x 2 ) = min { a x 1 , b x 2 } . f(x_1,x_2)=\min\{ax_1,bx_2\}. f(x1,x2)=min{ax1,bx2}.
c ( w 1 , w 2 , y ) = w 1 y a + w 2 y b = y ( w 1 a + w 2 b ) . c(w_1,w_2,y)=\frac{w_1y}{a}+\frac{w_2y}{b}=y\left(\frac{w_1}{a}+\frac{w_2}{b}\right). c(w1,w2,y)=aw1y+bw2y=y(aw1+bw2). -
CES技术
min w 1 x 1 + w 2 x 2 s.t. x 1 ρ + x 2 ρ = y ρ \min w_1x_1+w_2x_2 \\ \textbf{s.t. } x_1^{\rho} + x_2^{\rho} = y^{\rho} minw1x1+w2x2s.t. x1ρ+x2ρ=yρ
从 CES 生 产 函 数 推 导 成 本 函 数 {\color{red}从\text{CES}生产函数推导成本函数} 从CES生产函数推导成本函数:
第一步:从约束条件中得到 x 2 x_2 x2 的表达式
x 1 ρ + x 2 ρ = y ρ x 2 ρ = y ρ − x 1 ρ x 2 = ( y ρ − x 1 ρ ) 1 ρ x_1^{\rho} + x_2^{\rho} = y^{\rho} \\ x_2^{\rho} = y^{\rho} - x_1^{\rho} \\ x_2 = \left(y^{\rho} - x_1^{\rho} \right)^{\frac{1}{\rho}} \\ x1ρ+x2ρ=yρx2ρ=yρ−x1ρx2=(yρ−x1ρ)ρ1
第二步:将 x 2 x_2 x2 代入到优化函数中
min w 1 x 1 + w 2 ( y ρ − x 1 ρ ) 1 ρ \min w_1x_1 + w_2\left(y^{\rho} - x_1^{\rho} \right)^{\frac{1}{\rho}} minw1x1+w2(yρ−x1ρ)ρ1
第三步:对优化函数求导数,得极值点
令
t ( x 1 ) = w 1 x 1 + w 2 ( y ρ − x 1 ρ ) 1 ρ t(x_1) = w_1x_1 + w_2\left(y^{\rho} - x_1^{\rho} \right)^{\frac{1}{\rho}} t(x1)=w1x1+w2(yρ−x1ρ)ρ1
t ′ ( x 1 ) = w 1 + w 2 1 ρ ( y ρ − x 1 ρ ) 1 ρ − 1 ( − ρ x 1 ρ − 1 ) = w 1 − w 2 x 1 ρ − 1 ( y ρ − x 1 ρ ) 1 ρ − 1 t'(x_1) = w_1 + w_2\frac{1}{\rho}\left(y^{\rho}-x_1^{\rho} \right)^{\frac{1}{\rho}-1}\left(-\rho x_1^{\rho-1}\right)=w_1-w_2x_1^{\rho-1}\left(y^{\rho}-x_1^{\rho} \right)^{\frac{1}{\rho}-1} t′(x1)=w1+w2ρ1(yρ−x1ρ)ρ1−1(−ρx1ρ−1)=w1−w2x1ρ−1(yρ−x1ρ)ρ1−1
令 t ′ ( x 1 ) = 0 t'(x_1)=0 t′(x1)=0,得
w 1 − w 2 x 1 ρ − 1 ( y ρ − x 1 ρ ) 1 ρ − 1 = 0 w 1 = w 2 x 1 ρ − 1 1 ( y ρ − x 1 ρ ) ρ − 1 ρ w 1 ( y ρ − x 1 ρ ) ρ − 1 ρ = w 2 x 1 ρ − 1 w 1 ρ ρ − 1 ( y ρ − x 1 ρ ) = w 2 ρ ρ − 1 x 1 ρ w 1 ρ ρ − 1 x 1 ρ + w 2 ρ ρ − 1 x 1 ρ = w 1 ρ ρ − 1 y ρ x 1 ρ = w 1 ρ ρ − 1 w 1 ρ ρ − 1 + w 2 ρ ρ − 1 y ρ x 1 = [ w 1 ρ ρ − 1 w 1 ρ ρ − 1 + w 2 ρ ρ − 1 ] 1 ρ y . w_1-w_2x_1^{\rho-1}\left(y^{\rho}-x_1^{\rho} \right)^{\frac{1}{\rho}-1} = 0 \\ w_1 = w_2 x_1^{\rho-1} \frac{1}{\left(y^{\rho}-x_1^{\rho} \right)^{\frac{\rho-1}{\rho}}} \\ w_1\left(y^{\rho}-x_1^{\rho} \right)^{\frac{\rho-1}{\rho}} = w_2x_1^{\rho-1} \\ w_1^{\frac{\rho}{\rho-1}}\left(y^{\rho}-x_1^{\rho} \right) = w_2^{\frac{\rho}{\rho-1}} x_1^{\rho} \\ w_1^{\frac{\rho}{\rho-1}} x_1^{\rho} + w_2^{\frac{\rho}{\rho-1}} x_1^{\rho} = w_1^{\frac{\rho}{\rho-1}} y^{\rho} \\ x_1^{\rho} = \frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} y^{\rho} \\ x_1 = \left[\frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y. w1−w2x1ρ−1(yρ−x1ρ)ρ1−1=0w1=w2x1ρ−1(yρ−x1ρ)ρρ−11w1(yρ−x1ρ)ρρ−1=w2x1ρ−1w1ρ−1ρ(yρ−x1ρ)=w2ρ−1ρx1ρw1ρ−1ρx1ρ+w2ρ−1ρx1ρ=w1ρ−1ρyρx1ρ=w1ρ−1ρ+w2ρ−1ρw1ρ−1ρyρx1=[w1ρ−1ρ+w2ρ−1ρw1ρ−1ρ]ρ1y.
即
x 1 ( w 1 , w 2 , y ) = [ w 1 ρ ρ − 1 w 1 ρ ρ − 1 + w 2 ρ ρ − 1 ] 1 ρ y . x_1(w_1,w_2,y)= \left[\frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y. x1(w1,w2,y)=[w1ρ−1ρ+w2ρ−1ρw1ρ−1ρ]ρ1y.
第四步:将
x
1
x_1
x1 代入到
x
2
x_2
x2 的表达式中
x
2
=
(
y
ρ
−
x
1
ρ
)
1
ρ
=
(
y
ρ
−
[
[
w
1
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
1
ρ
y
]
ρ
)
1
ρ
=
[
y
ρ
−
w
1
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
y
ρ
]
1
ρ
=
y
[
w
2
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
.
\begin{aligned} x_2 &= \left(y^{\rho} - x_1^{\rho} \right)^{\frac{1}{\rho}} \\ &=\left(y^{\rho} - \left[ \left[\frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y \right]^{\rho} \right)^{\frac{1}{\rho}} \\ &= \left[y^{\rho}-\frac{w_1^\frac{\rho}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} y^{\rho} \right]^{\frac{1}{\rho}} \\ &= y\left[\frac{w_2^\frac{\rho}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} \right]. \end{aligned}
x2=(yρ−x1ρ)ρ1=⎝⎜⎛yρ−⎣⎡[w1ρ−1ρ+w2ρ−1ρw1ρ−1ρ]ρ1y⎦⎤ρ⎠⎟⎞ρ1=[yρ−w1ρ−1ρ+w2ρ−1ρw1ρ−1ρyρ]ρ1=y[w1ρ−1ρ+w2ρ−1ρw2ρ−1ρ].
即
x
2
(
w
1
,
w
2
,
y
)
=
y
[
w
2
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
.
x_2(w_1, w_2, y) = y\left[\frac{w_2^\frac{\rho}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} \right].
x2(w1,w2,y)=y[w1ρ−1ρ+w2ρ−1ρw2ρ−1ρ].
第五步:将
x
1
x_1
x1、
x
2
x_2
x2 代回成本函数中
c
(
w
1
,
w
2
,
y
)
=
w
1
x
1
(
w
1
,
w
2
,
y
)
+
w
2
x
2
(
w
1
,
w
2
,
y
)
=
w
1
[
w
1
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
1
ρ
y
+
w
2
[
w
2
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
y
=
[
w
1
ρ
w
1
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
1
ρ
y
+
[
w
2
w
2
ρ
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
y
=
[
w
1
ρ
2
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
1
ρ
y
+
[
w
2
2
ρ
−
1
ρ
−
1
w
1
ρ
ρ
−
1
+
w
2
ρ
ρ
−
1
]
y
.
\begin{aligned} c(w_1,w_2, y) &= w_1x_1(w_1,w_2,y)+w_2x_2(w_1,w_2,y) \\ &= w_1 \left[\frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y + w_2\left[\frac{w_2^\frac{\rho}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} \right]y \\ &= \left[w_1^{\rho}\frac{w_1^{\frac{\rho}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y + \left[w_2 \frac{w_2^\frac{\rho}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} \right]y \\ &= \left[\frac{w_1^{\frac{\rho^2}{\rho-1}}}{w_1^{\frac{\rho}{\rho-1}}+w_2^{\frac{\rho}{\rho-1}}} \right]^{\frac{1}{\rho}}y + \left[\frac{w_2^\frac{2\rho-1}{\rho-1}}{w_1^\frac{\rho}{\rho-1}+w_2^\frac{\rho}{\rho-1}} \right]y. \end{aligned}
c(w1,w2,y)=w1x1(w1,w2,y)+w2x2(w1,w2,y)=w1[w1ρ−1ρ+w2ρ−1ρw1ρ−1ρ]ρ1y+w2[w1ρ−1ρ+w2ρ−1ρw2ρ−1ρ]y=[w1ρw1ρ−1ρ+w2ρ−1ρw1ρ−1ρ]ρ1y+[w2w1ρ−1ρ+w2ρ−1ρw2ρ−1ρ]y=⎣⎡w1ρ−1ρ+w2ρ−1ρw1ρ−1ρ2⎦⎤ρ1y+⎣⎡w1ρ−1ρ+w2ρ−1ρw2ρ−12ρ−1⎦⎤y.
(
有
问
题
)
{\color{red}(有问题)}
(有问题)