Question about defn. of derivative

  • Context: Graduate 
  • Thread starter Thread starter AxiomOfChoice
  • Start date Start date
  • Tags Tags
    Derivative
Click For Summary

Discussion Overview

The discussion revolves around the definition of the derivative in calculus, specifically exploring the expression of the derivative as a limit of a sequence of functions. The scope includes theoretical aspects of calculus and mathematical reasoning.

Discussion Character

  • Exploratory
  • Mathematical reasoning

Main Points Raised

  • One participant proposes that if the derivative f'(x) exists, it can be expressed as the limit of a sequence of functions defined by f_n(x) = (f(x+h_n) - f(x))/h_n, where h_n is a sequence converging to 0.
  • Another participant questions the need for uniform convergence in the context of the derivative's definition.
  • A later reply affirms the initial proposal but suggests that it is more cumbersome than the standard definition of the derivative.
  • One participant introduces the equivalence of limits, stating that the limit of f(x) as x approaches a equals L if and only if the limit of f(a_n) as n approaches infinity equals L for every sequence converging to a.

Areas of Agreement / Disagreement

Participants generally agree on the validity of the proposed expression for the derivative, but there is uncertainty regarding the necessity of uniform convergence and the preference for the standard definition. The discussion remains unresolved on these points.

Contextual Notes

There is a lack of clarity regarding the implications of uniform convergence in the context of the derivative, and the discussion does not resolve the conditions under which the proposed limit holds.

AxiomOfChoice
Messages
531
Reaction score
1
Let [itex]\{h_n\}[/itex] be ANY sequence of real numbers such that [itex]h_n \neq 0[/tex] and [itex]h_n \to 0[/itex]. If [itex]f'(x)[/itex] exists, do we have<br /> <br /> [tex] f'(x) = \lim_{n\to \infty} f_n(x),[/tex]<br /> <br /> where <br /> [tex] f_n(x) = \frac{1}{h_n} (f(x+h_n) - f(x))[/tex]<br /> <br /> ?<br /> <br /> This seems to express the derivative as the pointwise limit of a sequence of functions...right? Do we know, in addition, that the convergence is uniform?[/itex]
 
Physics news on Phys.org
Uniform in what sense? The definition does not require uniformity in x.
 
mathman said:
Uniform in what sense? The definition does not require uniformity in x.

Ok, I'm not sure :) I didn't really think before writing out that question. But am I right on all other counts?
 
You are right, but it is more cumbersome than the usual approach where:

f'(x)=lim(h->0) (f(x+h) - f(x))/h
 
[itex]lim_{x\to a} f(x)= L[/itex] if and only if [itex]\lim_{n\to \infty} f(a_n)= L[/itex] for every sequence [itex]\{a_n\}[/itex] that converges to a. The two formulations are equivalent.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K