Multivariable factor decomposition

Click For Summary
SUMMARY

The discussion centers on the process of multivariable factor decomposition for the expression f(x,y)=\frac{y^2-xy+1}{(x+y)(x-y)}. The initial attempt to decompose this expression into the form \frac{A}{x+y} + \frac{B}{x-y} is flawed, as it leads to an inconsistency in the equations derived. Participants clarify that the decomposition cannot yield valid constants A and B for all x and y, and suggest alternative methods involving specific substitutions to find A and B.

PREREQUISITES
  • Understanding of multivariable calculus, specifically partial derivatives.
  • Familiarity with algebraic manipulation and factor decomposition techniques.
  • Knowledge of rational functions and their properties.
  • Experience with solving systems of equations.
NEXT STEPS
  • Study the method of partial fraction decomposition in multivariable calculus.
  • Learn about the implications of substituting specific values in algebraic expressions.
  • Explore the use of partial derivatives in optimization problems.
  • Investigate alternative approaches to factor decomposition in rational functions.
USEFUL FOR

Mathematicians, students of calculus, and anyone involved in algebraic manipulation and factor decomposition in multivariable contexts will benefit from this discussion.

epkid08
Messages
264
Reaction score
1
(Keep in mind, I made this off the top of my head, so if something cancels easy, ignore it)

Let's say I had this expression:

[tex]f(x,y)=\frac{y^2-xy+1}{(x+y)(x-y)}[/tex]

I want to decompose this to:

[tex]\frac{A}{x+y} + \frac{B}{x-y}[/tex]

So i begin the process:
[tex]y^2-xy+1=A(x-y) + B(x+y)[/tex]
[tex]y^2-xy+1=x(A+B) + y(B-A)[/tex]At this point, I can't just plug in random numbers, I need to plug in points that correspond to this point:
[tex](\frac{\partial f}{\partial x}, \frac{\partial f}{\partial y})[/tex]

So I would pick a point, say (1,1), plug it into [tex]\frac{\partial f}{\partial x}[/tex] to get the x coordinate of a point, [tex]P_1[/tex], then I would plug in the same point, (1,1), into [tex]\frac{\partial f}{\partial y}[/tex] to get the y coordinate of the [tex]P_1[/tex]. After I get two points,[tex]P_1[/tex] and [tex]P_2[/tex], I can plug them into the equation and solve the system to find A and B.

My question is, is this the right process to decompose the fraction, or am I way off?
 
Physics news on Phys.org
epkid08 said:
Let's say I had this expression:

[tex]f(x,y)=\frac{y^2-xy+1}{(x+y)(x-y)}[/tex]

I want to decompose this to:

[tex]\frac{A}{x+y} + \frac{B}{x-y}[/tex]

So i begin the process:
[tex]y^2-xy+1=A(x-y) + B(x+y)[/tex]
[tex]y^2-xy+1=x(A+B) + y(B-A)[/tex]

At this point you see a problem. There does not exist such numbers A and B that the last equation would be true for all x and y. It follows that your decomposition attempt doesn't work.

At this point, I can't just plug in random numbers, I need to plug in points that correspond to this point:
[tex](\frac{\partial f}{\partial x}, \frac{\partial f}{\partial y})[/tex]

What points are you talking about?

My question is, is this the right process to decompose the fraction, or am I way off?

I did not understand your process, but since the task you are trying to do is impossible, I might guess your process was not right.
 
epkid08 said:
So i begin the process:
[tex]y^2-xy+1=A(x-y) + B(x+y)[/tex]
[tex]y^2-xy+1=x(A+B) + y(B-A)[/tex]

Why are you messing with partials? Let x = y in the first expression and you get

[tex]x^2 - x^2 + 1 = 2Bx \Rightarrow B = \frac{1}{2x}[/tex]

Similarly let x = -y and you get

[tex]x^2 + x^2 + 1 = 2Ax \Rightarrow A = x^{2} + \frac{1}{2}[/tex]
 

Similar threads

  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 16 ·
Replies
16
Views
3K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K