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: Differentiation question

  1. Nov 16, 2007 #1
    Problem statement

    Show that if F:(a,b)->R is differentiable at every point of (a,b) and increasing then f'(x) >= 0 for all x belonging to (a,b)

    Problem Solution

    Definition of differentiation:
    Assume sequence {an} -> a then
    f(an) - f(x) / (an -x) converges to some value

    Using this definition I am saying since it is known that f is differentiable at a,b then the following is true

    Assume sequence an->a and bn->b
    then f(bn) -f(an) / (bn-an) --- (eq1)

    Since this is an increasing function both numerator and denominator will be always greater than 0 and thus the derivative by eq1 is always positive.

    Am I on the right track?

    If not plz advise.


  2. jcsd
  3. Nov 17, 2007 #2


    User Avatar
    Science Advisor
    Homework Helper

    I think you meant [f(b) -f(bn)] / (b-bn) and alternatively [f(an) -f(a)] / (an-a). Even then this only addresses f'(a) > 0 and f'(b) > 0.
  4. Nov 17, 2007 #3
    No I meant what I had written - however I did realise after posting that it was wrong. I worked through a specific f(x) and figured it was wrong.

    If I use the definition of differentiability as you have done (f(b) - f(bn))/(b-bn), can I safely say that it is f'(b) >= 0;
  5. Nov 17, 2007 #4


    User Avatar
    Science Advisor

    Is that really the definition of "differentiation" your text gives?

    It should be "f is differentiable at a if and only if for every sequence {an} that converges to a, (f(an)- f(a))/(an-a) converges to the same limit. If f is differentiable at a, then its derivative is that common limit."
    (NOT "f(an) - f(x) / (an -x)" that makes no sense. You must mean (f(an)- f(a))/(an-a) with "a", not "x". )

    The way you've written it, it sounds like just one such sequence would be sufficient to determine that a function is differentiable. Of course, here, since you are given that f is differentiable, then it really doesn't matter.

    I see no reason for looking at two sequences {an} and {bn}. Take {an} to be a sequence converging to a with an> a for all n.
    Last edited by a moderator: Nov 17, 2007
  6. Nov 17, 2007 #5
    Just use mean-value theorem.
  7. Nov 17, 2007 #6


    User Avatar
    Science Advisor

    Well, if you want to do things the easy way!! :)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook