Max and Min of function via Lagrange multipliers

Click For Summary

Homework Help Overview

The problem involves finding the maximum and minimum values of the function \( f(x,y) = (x-y)^n \) under the constraint \( g(x,y) = x^2 + 3y^2 - 1 = 0 \), where \( n \) is a fixed positive integer. Participants are exploring the application of Lagrange multipliers in this context.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • Participants discuss identifying critical points and the implications of having multiple values for \( x \) with fewer equations. There is mention of splitting the analysis based on whether \( n \) is even or odd, and questions arise regarding the correctness of this approach.

Discussion Status

The discussion is ongoing, with participants providing attempts at solutions and seeking clarification on their methods. Some guidance has been offered regarding the nature of equations and roots, but no consensus has been reached on the overall approach or correctness of the methods used.

Contextual Notes

Participants note the challenge of proving cases for different values of \( n \) and express uncertainty about whether their methods align with the expectations of the problem. There is also a mention of the need for more detailed calculations to facilitate feedback.

CAF123
Gold Member
Messages
2,918
Reaction score
87

Homework Statement


If n is a fixed positive integer, compute the max and min values of the function (x-y)^n = f(x,y), under the constraint x^2 + 3y^2 = 1

The Attempt at a Solution


I got the 4 critical points (±\frac{\sqrt{3}}{2}, ±\frac{1}{2\sqrt{3}})\,\,\text{and}\,\, (±\frac{1}{2},±\frac{1}{2})
Correct? My question is : to find the max and min of this function under this constraint, I split n into even and odd cases and found the values from there. Is that the right way to go about the question?
 
Physics news on Phys.org
Can anyone offer any advice?
I alo noted that I got 4 values of x but I only had 3 equations. Does this make sense?
 
I think you would get more feedback if you actually show your calculations instead of just your answer.
 
CAF123 said:
Can anyone offer any advice?
I alo noted that I got 4 values of x but I only had 3 equations. Does this make sense?

Sure: you can get more than one value from a single equation. For example, the single equation x^2 = 4 has two roots: x = 2 and x = -2.

RGV
 
micromass said:
I think you would get more feedback if you actually show your calculations instead of just your answer.
I computed f_{x}, f_{y}, g_{x}, g_{y} and solved the system of equations: n(x-y)^{n-1} = 2\lambda x, -n(x-y)^{n-1} = 6\lambda y\,\,\text{and}\,\, x^2 + 3y^2 =1 Adding the first of these equations gives λ=0 or x=-3y. Sub x=-3y into the constraint to give y= ±1/sqrt(12) which implies x= ±(sqrt(3))/2. Now using λ=0 gives x=y , which again by the constraint gives x=±1/2 and y=±1/2. Correct?

To find the max and min I split n into even and odd. I am pretty confident this is correct because I have suitable answers, but I was wondering if the question wanted me to do something else(purely because I had to prove the cases for any n=2k, n= 2k+1, etc.. which that sort of style of working I had not done before in calculus)
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 16 ·
Replies
16
Views
3K
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 15 ·
Replies
15
Views
4K
  • · Replies 18 ·
Replies
18
Views
3K