Delta Epsilon Limit Proof - f(x,y) = xy

Click For Summary

Discussion Overview

The discussion centers on proving the limit of the function f(x,y) = xy as (x,y) approaches (a,b) using the precise definition of a limit. Participants explore various approaches to establish the limit, engaging with the epsilon-delta definition in a multi-variable context.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant suggests starting with the expression |xy - ab| < ε and proposes conditions on |x - a| and |y - b| to derive bounds for xy.
  • Another participant provides an alternative approach by rewriting |xy - ab| and using the triangle inequality, indicating that |y| can be bounded as y approaches b.
  • A different participant raises a concern about the output of LaTeX formatting in the discussion.
  • One participant assumes a and b are non-zero and presents a method involving inequalities to relate |xy - ab| to ε, suggesting a summation of inequalities.
  • Another participant proposes a multiplication approach, attempting to establish bounds but later questions the validity of their reasoning due to a sign error.

Areas of Agreement / Disagreement

Participants present multiple competing views and approaches to the limit proof, with no consensus reached on a single method or solution. The discussion remains unresolved as various strategies are explored.

Contextual Notes

Some participants express uncertainty about the correctness of their approaches, and there are indications of missing assumptions or dependencies on the values of a and b. The discussion also highlights potential issues with mathematical expressions and clarity in reasoning.

JeffNYC
Messages
26
Reaction score
0
How does one prove:

limit xy = ab
x-> a
y -> b

using the precise definition of a limit?

My attempt:

|xy-ab|<ϵ

for:

0<|x-a|<δ/2
0<|y-b|<δ/2

it follows that:

δ/2-a <x< δ/2+a
δ/2-b <y< δ/2+b

then:

(δ/2-a)(δ/2-n) < xy < (δ/2+a)(δ/2+b)

(δ^2/4-aδ/2-bδ/2 +ab) <xy< (δ^2/4+aδ/2+bδ/2+ab )

So -

|xy-ab|< (δ^2/4+aδ/2+bδ/2)

and I need to make

|xy-ab|<ϵ

At this point I feel a bit lost - anyone have any ideas for this proof?

Thanks,

Jeff
 
Physics news on Phys.org
Yes, you want |xy- ab|&lt; \epsilon . Now, xy- ab= xy- ay+ ay- ab= y(x-a)+ a(y-b) so |xy- ab|\le |y||x-a|+ |a||y- b|. Since y converges to b, you can make |y-b| as small as you like so you can make it less than \epsilon/2a. The other term is a bit tricky. You need to make |y||x-a| less than \epsilon/2 but y is a variable. However, since y converges to b, you can make |y| less than |b|+1 so you can use that instead: since x converges to a, you can make |x-a|\le \epsilon/(|b|+ 1)&lt; \epsilon/|y|.
 
Any idea what's happening with the latex output?
 
( I assume a,b ≠ 0 )

| x - a | < ε' hence | x∙y - a∙y | < ε' ∙ | y |
| y - b | < ε' hence | y∙a - b∙a | < ε' ∙ | a |

you sum both inequalities

| x∙y - a∙y | + | y∙a - b∙a | < ε' ∙ | y | + ε' ∙ | a |

you use the triangle inequality

| x∙y - a∙y | + | y∙a - b∙a | > | x∙y - a∙y + y∙a - b∙a | = | x∙y - a∙b |

define ε = ε' ∙ | y | + ε' ∙ | b |

now putting everything together yields : ...
 
Last edited:
maybe multiplying works too?

-d1 + a < x < d1 + a
-d2 + b < y < d2 + b

=>
since -d1d2 < d1d2,

-d1d2-d1b-d2a < d1d2-d1b-d2a < xy - ab < d1d2 + d1b + d2a

then,

abs(xy - ab) < d1d2 + d1b + d2a

since a and b are fixed, as d1 and d2 go to zero, the abs goes to zero.

[edit: hm. no, the sign on the multiplication makes this an error.]
 
Last edited:

Similar threads

  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 11 ·
Replies
11
Views
6K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 28 ·
Replies
28
Views
4K
  • · Replies 9 ·
Replies
9
Views
2K