Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
0% found this document useful (0 votes)
16 views2 pages

Lagrange Multipliers 2

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 2

Lagrange Multipliers 2

This is a follow on sheet to Lagrange Multipliers 1 and as promised, in this sheet we will look at
an example in which the Lagrange multiplier λ has a concrete meaning and this will enable us to
find the answer to a related optimization problem without having to go through the whole process
of solving the Lagrange equations again.

Cobb-Douglas Production Functions

Let q denote the quantity produced of a good. In general this will depend on the amount of capital
and labour employed in the production. As a first approximation, we will assume that

q = f (K, L),

where K denotes capital and L denotes labour.

A Cobb-Douglas production function relates the quantities q, K , and L in the following way:

q = cK α Lβ ,

where α, β and c are constants and α and β are such that 0 < α < 1 and 0 < β < 1.

Example

Consider the Cobb-Douglas production function

q = 30x2/3 y 3/10 ,

where x represents the number of units of capital and y represents the number of units of labour.
Suppose that a firm’s unit capital and labour costs are e5 and e6 respectively.

1. Find the values of x and y that maximise output if the total input costs are fixed at e7250.

2. Find the new maximum output if the input costs are increased to e7300.
Solution
2 3
1. We have to maximize q = 30x 3 y 10 subject to 5x + 6y = 7250.

We let our constraint equation be g(x) = 5x + 6y − 7250 = 0.

Since ∇g = (5, 6) 6= (0, 0), there exists a λ ∈ R such that ∇q = λ∇g .

We need to solve ∇q = λ∇g and g = 0. Now


3 2
!
 1 3 2 7
 20y 10 9x 3
∇q = 20x− 3 y 10 , 9x 3 y − 10 = 1 , 7 = λ(5, 6).
x3 y 10

Thus 3 2
20y 10 9x 3
1 = 5λ and 7 = 6λ.
x3 y 10
Hence 3 2
4y 10 3x 3 3
λ= 1 = 7 =⇒ 8y = 3x =⇒ y = x.
x 3 2y 10 8

Substituting this in the constraint equation, we obtain

3 29
5x + 6 · x = 7250 =⇒ x = 7250 =⇒ x = 1000 and then y = 375.
8 4
We now check the value of q at (1000, 375) and also at the endpoints of the constraint line
5x + 6y = 7250 to determine where the maximum occurs. Since we are only interested in
non-negative values of x and y , the endpoints of the constraint line 5x + 6y = 7250 lie where
where it cuts the x and y axes, that is at (1450, 0) and 0, 3625
3
.

Now  
3625
q(1450, 0) = q 0, = 0 and q(1000, 375) ≃ 17755.
3
Thus the maximum is indeed attained at x = 1000, y = 375.

2. We will use the fact that if the input costs are increased by e1 then the maximum output
is increased by the Lagrange multiplier λ, which in this context is called The Marginal
Productivity of Money.
So in this case the marginal productivity of money is

3
4(375) 10
λ= 1 ≃ 2.37.
(1000) 3
Since we are increasing the input costs by e50, the new maximum output is the old maximum
output plus 50λ, so it is
q(1000, 375) + 50λ ≃ 17873.

You might also like