1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: General Epsilon-Delta Proof

  1. Aug 19, 2011 #1
    1. The problem statement, all variables and given/known data

    Hey guys, I've been given the following [itex] \epsilon - \delta [/itex] proof question. The trouble I'm having is I'm not 100% sure what it is actually asking for, and how to go about it. The more common proofs involving two variable functions and a given limit are easy enough to do, but I'm stuck on this one.

    2. Relevant equations

    Suppose [itex] f: \left( \mathbb R ^2 , \| . \|_2 \right) \rightarrow \left( \mathbb R , | .| \right) [/itex] is continuous at [itex] \left(a,b \right) \in \mathbb R^2 [/itex]. Prove using the [itex] \epsilon - \delta [/itex] definition only, that if we define the function

    [itex] f_b : \left( \mathbb R,|.| \right) \rightarrow \left( \mathbb R,|.| \right) [/itex] s.t. [itex] f_b \left(x \right) = f \left( x,b \right), [/itex]

    then [itex] f_b [/itex] is continuous at [itex] x=a [/itex].

    3. The attempt at a solution

    Well, considering I am not actually sure what the question is acting for I can't go very far. I was assuming you would set up the usual proof as,

    Let [itex] \epsilon > 0 [/itex]
    if [itex] |(x,b) - (a,b)| < \delta [/itex]
    then [itex] |f(x,b) - (a,b)| < \epsilon [/itex]

    Possibly then, [itex] x - a < \delta [/itex]?

    Any help would be appreciated, cheers.
    Last edited: Aug 19, 2011
  2. jcsd
  3. Aug 19, 2011 #2
    Take [itex]\varepsilon >0[/itex] fixed.

    You know that f is continuous at (a,b), thus you know that there exists a [itex]\delta>0[/itex] such that for all (x,y) it holds that


    Now, what you must do is to show [itex]f_b[/itex] continuous at a. Thus you must find a [itex]\delta>0[/itex] such that for all x it holds that


    Can you proceed now?
  4. Aug 19, 2012 #3
    I am also interested in this question. I understand what is required for the proof, but I am just stuck as to how to manipulate:
    |f(x,b) - f(a,b)|
    into something that resembles δ??

    We know that:
    0 < ||(x,y) - (a,b) || < δ [itex]\Rightarrow[/itex] |f(x,y) - f(a,b)| < ε
    as the limit is equal to the value of the function at (a,b) since it is continuous there, but how can we follow on from this to show that if y is restricted to equalling b, then:
    0 < ||(x,b) - (a,b) || = ||x-a|| < δ [itex]\Rightarrow[/itex] |f(x,b) - f(a,b)| < ε
    ?? I'm really stuck and any help would be appreciated.
  5. Aug 19, 2012 #4


    User Avatar
    Science Advisor

    How is ||(x, y)- (a, b)|| defined? (there are several equivalent definitions- which are you using?)
  6. Aug 20, 2012 #5
    ||(x,y) - (a,b)|| is a norm measuring the distance from (a,b) to (x,y), what do you mean how is it defined? Distance between points in R^2, euclidean norm.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook