Scribd
Upload
15 views

Lect 21

This document discusses how to use the second derivative test and method of Lagrange multipliers to find constrained critical points of multivariate functions. It provides an example of using Lagrange multipliers to find the constrained critical points of a function of three variables subject to a single constraint.

Uploaded by

adityastkhsg
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
15 views

Lect 21

This document discusses how to use the second derivative test and method of Lagrange multipliers to find constrained critical points of multivariate functions. It provides an example of using Lagrange multipliers to find the constrained critical points of a function of three variables subject to a single constraint.

Uploaded by

adityastkhsg
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

We now detail how the second derivative test works in the case of 3

variables.

Let w = f (x, y, z) have continuous second partial derivatives near a

critical point (a, b, c).

The Hessian matrix is  


 fxx(a, b, c) fxy (a, b, c) fxz (a, b, c) 
 
 
A = H f (a, b, c) =  fxy (a, b, c) .
 
 fyy (a, b, c) fyz (a, b, c) 

 
 
fxz (a, b, c) fyz (a, b, c) fzz (a, b, c)
We first compute det A = det(H f ). If det A 6= 0, we can proceed.
 
If det A1 = det fxx(a, b, c) > 0
 
 fxx(a, b, c) fxy (a, b, c) 
det A2 = det  >0
 
 
fxy (a, b, c) fyy (a, b, c)
det A3 = det A > 0

we have the sequence + + + which indicates a relative minimum.


 
If det A1 = det fxx(a, b, c) < 0
 
 fxx(a, b, c) fxy (a, b, c) 
det A2 = det  >0
 
 
fxy (a, b, c) fyy (a, b, c)
det A3 = det A < 0

we have the sequence − + − which indicates a relative maximum.

If we have any other sequence, the critical point is neither a minimum

nor a maximum.

If det A = 0, we have the degenerate case and we can not determine

if the critical point is a maximum, a minimum or neither from the

second derivative test.


Suppose we have a function f (x, y, z) subject to the constraint

g(x, y, z) = 0. We define a new function F of 4 variables by

F (x, y, z, λ) = f (x, y, z) − λ (g(x, y, z))

If (a, b, c, λ0) is a critical point of F , then (a, b, c) is a critical point

of f subject to the constraint g(x, y, z) = 0.

The λ is called a Lagrange multiplier and the new function F

we defined is called the Lagrangian or the Lagrange function.

This approach to solving for (constrained) critical points is called the

method of Lagrange multipliers.

(Although stated for 3 variables, this approach can easily be adapted

for 2, 4, 5 or more variables.)


Suppose we have a function f (x, y, z) subject to the constraint

g(x, y, z) = 0. We define a new function F of 4 variables by

F (x, y, z, λ) = f (x, y, z) − λ (g(x, y, z))

If (a, b, c, λ0) is a critical point of F , then (a, b, c) is a critical point

of f subject to the constraint g(x, y, z) = 0.

Find the constrained critical points of

f (x, y, z) = (x − 3)2 + (y − 1)2 + (z + 1)2

subject to the constraint

g(x, y, z) = x2 + y 2 + z 2 − 4 = 0 .

We define the Lagrangian

F (x, y, z, λ) = f (x, y, z) − λ g(x, y, z)

= (x − 3)2 + (y − 1)2 + (z + 1)2 − λ (x2 + y 2 + z 2 − 4)

and calculate the partials and equate to 0:


Fx = 2(x − 3) − 2λx =0.

Fy = 2(y − 1) − 2λy =0

Fz = 2(z + 1) − 2λz =0

Fλ = −(x2 + y 2 + z 2 − 4) = 0
We will solve the first 3 in terms of λ and then substitute into the

4th.
3
From the first: x − 3 = λ x =⇒ x(1 − λ) = 3 =⇒ x =
1−λ
We note that 1−λ 6= 0 because, if λ = 1, we would have 2x−6−2x =

0 =⇒ −6 = 0 which is impossible.
1
From the second: y − 1 = λ y =⇒ y(1 − λ) = 1 =⇒ y =
1−λ
−1
From the third: z + 1 = λ z =⇒ z(1 − λ) = −1 =⇒ z = .
1−λ
32 12 (−1)2
Now the fourth becomes + + = 4 =⇒
(1 − λ)2 (1 − λ)2 (1 −√λ)2
11 2 11 11
= 4 =⇒ (1 − λ) = =⇒ 1 − λ = ± =⇒ λ =
(1 −√λ)2 4 2
11
1±
2
3 6
Hence x =   = ±√
√
1 − 1 ± 211 11
1 2
y=  √  = ± √
1 − 1 ± 11 11
2
−1 2
z=   = ∓√ .
√
1 − 1 ± 211 11
We have 2 (constrainted) critical points
   
6 2 −2 −6 −2 2
√ , √ , √ and √ , √ , √ .
11 11 11 11 11 11

You might also like