Scribd
Upload
41 views2 pages

Implicit Function Theorem: Exam 2014, Problem 4 (B) and (C)

The document discusses applying the implicit function theorem and envelope theorem to solve optimization problems involving parameters. It presents an optimization problem with objective function U depending on parameters θ1 and θ2 and variables x1 and x2. The implicit function theorem is used to find the derivatives of the optimal values x1* and x2* with respect to the parameters. It provides the calculations to show that dx2*/dθ2 is positive and dx1*/dθ2 depends on the sign of f12. The envelope theorem is then introduced, stating that the derivative of the maximum value function M with respect to a parameter equals the derivative of the objective U with respect to that parameter, evaluated at the optimal values. It

Uploaded by

Ziggy X Polke
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
41 views2 pages

Implicit Function Theorem: Exam 2014, Problem 4 (B) and (C)

The document discusses applying the implicit function theorem and envelope theorem to solve optimization problems involving parameters. It presents an optimization problem with objective function U depending on parameters θ1 and θ2 and variables x1 and x2. The implicit function theorem is used to find the derivatives of the optimal values x1* and x2* with respect to the parameters. It provides the calculations to show that dx2*/dθ2 is positive and dx1*/dθ2 depends on the sign of f12. The envelope theorem is then introduced, stating that the derivative of the maximum value function M with respect to a parameter equals the derivative of the objective U with respect to that parameter, evaluated at the optimal values. It

Uploaded by

Ziggy X Polke
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Implicit function theorem

For the general description of implicit function theorem, see Chapter 5, in particular slides 82-86.

Exam 2014, problem 4(b) and (c)


U (x1 , x2 ) = 1 f (x1 , x2 ) + 2 g(x2 ) c(x1 ) k(x2 )
First-order conditions:

(
1 f1 c1 = 0
1 f2 + 2 g2 k2 = 0.

Second-order conditions (follows from negative definiteness of the Hessian of U ):


(
1 f11 c11 < 0
2
(1 f11 c11 )(1 f22 + 2 g22 k22 ) 12 f12
> 0.

(1)

(2)

x1
x2
in point (b) and
in (c). Since x1 and x2 solve FOC, consider (1)
2
2
as a pair of functions, which depend on variables x1 , x2 and parameters 1 , 2 (as on slide 5-83):
(
F 1 (x1 , x2 , 1 , 2 ) = 1 f1 c1 = 0
(3)
F 2 (x1 , x2 , 1 , 2 ) = 1 f2 + 2 g2 k2 = 0.
You are looking for

To use implicit function theorem, you need to construct several matrices. The first one, D, is
the matrix of first-order derivatives of F i with respect to xj (slide 5-84). By the nature of this
problem, D is equivalent to the Hessian of U :

 

F 1 /x1 F 1 /x2
1 f11 c11
1 f12
D=
=
= HU .
F 2 /x1 F 2 /x2
1 f12
1 f22 + 2 g22 k22
e consists of minus first-order derivatives of F i with respect to j
Another matrix you need, D,
(slide 5-85):

 

F 1 /1 F 1 /2
f1 0
e
D=
=
.
F 2 /1 F 2 /2
f2 g2
x2
|Dx2 2 |
=
, where matrix Dx2 2 is obtained
2
|D|
from D by replacing the second column of D (it corresponds to the derivatives w.r.t. x2 ) by the
e (it corresponds to the derivatives w.r.t. 2 ):
second column of D


1 f11 c11 0
Dx2 2 =
.
1 f12
g2
Finally, implicit function theorem states that

The determinant of this matrix is equal to g2 (1 f11 c11 ). The determinant of D is the
determinant of the Hessian of U . Thus, the answer to point (b) is
x2
g2 (1 f11 c11 )
=
.
2
2
(1 f11 c11 )(1 f22 + 2 g22 k22 ) 12 f12
1

You can say something about the sign of this effect. From SOC you know that the denominator
is positive, and the term 1 f11 c11 in the numerator is negative. By assumption, g2 > 0, so the
x2
> 0.
whole fraction is positive:
2
x1
|Dx1 2 |
Point (c) is solved analogically,
=
. Now you need to write down matrix Dx1 2
2
|D|
obtained from D by replacing the first column of D (it corresponds to the derivatives w.r.t. x1 )
e (it corresponds to the derivatives w.r.t. 2 ):
by the second column of D


0
1 f12
Dx1 2 =
.
g2 1 f22 + 2 g22 k22
The determinant of this matrix is equal to 1 g2 f12 . The answer to point (c) is
x2
1 g2 f12
.
=
2
2
(1 f11 c11 )(1 f22 + 2 g22 k22 ) 12 f12
As before, the denominator is positive, 1 and g2 are positive, so the sign of this partial derivative
x
depends on the sign of f12 , which is unknown. You can only say that 2 > 0 if and only if f12 > 0.
2
(In other words, to determine the sign of the effect, you have to make an assumption about the
sign of f12 .)

Envelope theorem
For the reference, see Chapter 6, theorem 6.14.

Exam 2014, problem 4(d)


How does function U evaluated at x1 , x2 (aka maximum-value function) change with 2 ? Denote
the maximum-value function as M (1 , 2 ) = U (x1 (1 , 2 ), x2 (1 , 2 )). Envelope theorem states the
following (here we have an optimization problem without constraints):
M
U
(1 , 2 ) =
(x , x ).
2
2 1 2
To determine the effect of a parameter on the objective function at optimal values, it is enough
to take the derivative of the function w.r.t. the parameter and to plug in the optimal values
afterwards. In our case,




M


(1 , 2 ) =
(1 f (x1 , x2 ) + 2 g(x2 ) c(x1 ) k(x2 ))
= g(x2 )
= g(x2 ).
2
2

(x ,x )
(x ,x )
1

You might also like