Proving the Product of Jacobians Using the Chain Rule

  • Thread starter Thread starter Lucy Yeats
  • Start date Start date
  • Tags Tags
    Product Proof
Click For Summary

Homework Help Overview

The problem involves proving that the product of two Jacobians, A and B, corresponding to the transformations between two pairs of variables (x, y) and (u, v), equals the unit matrix I using the chain rule.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the multiplication of the Jacobians and the simplification of the resulting expressions. There are questions about the correctness of the partial derivatives and the interpretation of the results.

Discussion Status

Some participants have provided guidance on re-evaluating the simplifications made in the calculations. There is an ongoing exploration of the relationships between the variables and their derivatives, with no explicit consensus reached yet.

Contextual Notes

Participants are considering the implications of the variables being independent of each other and how that affects the partial derivatives involved in the Jacobians.

Lucy Yeats
Messages
117
Reaction score
0

Homework Statement



The pair of variables (x, y) are each functions of the pair of variables (u, v) and vice versa.
Consider the Jacobians A=d(x,y)/d(u,v) and B=d(u,v)/d(x,y). Show using the chain rule that the product AB of these two matrices equals the unit matrix I.

Homework Equations





The Attempt at a Solution



I wrote out the two Jacobians and tried to multiply them but I got the following:
(dx/du)(du/dx)+(dx/dv)(dv/dx) (dx/du)(du/dy)+(dx/dv)(dv/dy)
(dy/du)(du/dx)+(dy/dv)(dv/dx) (dy/du)(du/dy)+(dy/dv)(dv/dy)

= 2 2dy/dx
2dy/dx 2

Where did I go wrong/ how do I continue this question?
 
Physics news on Phys.org
Your product of A and B is correct, but I don't quite see how you got the result after that. Can you explain "= 2 2dy/dx " ?
2dy/dx 2

Perhaps you mean that those are what you get in the diagonals? I'm not sure.
 
I see what you were getting at. You're close enough to the solution, re-work the part where you simplified the partial derivatives. Check your differentials on the off-diagonal.

Edit: I removed the part about a constant multiplying the identity. Indeed B should be the inverse mapping that takes (u,v) back to (x,y), or the other way if you wish.
 
Last edited:
Sorry, I wasn't sure how to format matrices so it's a bit unclear.
For the first row, first column: (dx/du)(du/dx)+(dx/dv)(dv/dx)=1+1=2
First row, second column: (dx/du)(du/dy)+(dx/dv)(dv/dy)= (dx/dy)+(dx/dy)=2(dx/dy) as the u's and v's cancel out.
Second row, first column:(dy/du)(du/dx)+(dy/dv)(dv/dx)=(dy/dx)+(dy/dx)=2(dy/dx) as the u's and v's cancel out.
Second row, second column:(dy/du)(du/dy)+(dy/dv)(dv/dy)=1+1=2

I'm not sure where I've made a mistake...

Thanks so much for helping! :-)
 
I understand the terms on the main diagonal now, by manipulating the total derivative, but I'm still confused about the other two terms.

Any help would be great.
 
if y isn't a function of x at all, and x isn't a function of y at all, what can you say about ∂y/∂x and ∂x/∂y? think of the simplest example:

x(u,v) = u
y(u,v) = v

(so that u(x,y) = x, v(x,y) = y).
 
If y and x don't depend on each other, would the partial derivatives be zero?
 

Similar threads

Replies
19
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
6
Views
2K
Replies
21
Views
2K
  • · Replies 28 ·
Replies
28
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K