My Math Forum A new beautiful result on convex functions

 Topology Topology Math Forum

 June 19th, 2017, 11:57 AM #11 Senior Member   Joined: Mar 2015 From: New Jersey Posts: 1,175 Thanks: 90 Ascoli Theorem â€œBy definition, a sequence {â€‰fnâ€‰}nâˆˆN of continuous functions on an interval I = [a, b] is uniformly bounded if there is a number M such that |fn(x)| $\displaystyle \leq$ M for every function â€‰fnâ€‰ belonging to the sequence, and every x âˆˆ [a, b]. (Here, M must be independent of n and x.) The sequence is said to be equicontinuous if, for every Îµ > 0, there exists Î´ > 0 such that 1) |fn(x)-fn(y)| < $\displaystyle \epsilon$ whenever |x âˆ’ y| < Î´â€‰ for all functions â€‰fnâ€‰ in the sequence. (Here, Î´ may depend on Îµ but not x, y or n.) One version of the theorem can be stated as follows: Consider a sequence of real-valued continuous functions {â€‰fnâ€‰}n âˆˆ N defined on a closed and bounded interval [a, b] of the real line. If this sequence is uniformly bounded and equicontinuous, then there exists a subsequence {â€‰fnkâ€‰}k âˆˆ N that converges uniformly.- wicki So you are showing that your convex functions satiefy the conditions of Ascoli theorem. Why didnâ€™t you say so?. Originally you said fn(x) was defined on R and result applied for all R. As for your proof, your inequalities use that a convex function on [a,b] has a monotonically increasing or decrasing slope, which implies 1) if all the slopes at a and b are bounded. But you donâ€™t know that.
 June 19th, 2017, 12:20 PM #12 Member   Joined: May 2017 From: France Posts: 57 Thanks: 1 You applied this rÃ©sult for [-n,n] with $\displaystyle n \in \mathbb N$ "Why didnâ€™t you say so?." Your question, is it a joke ? It is the very principle of an enigma, we don't give the solution, with the enigma.
June 20th, 2017, 08:57 AM   #13
Senior Member

Joined: Mar 2015
From: New Jersey

Posts: 1,175
Thanks: 90

Quote:
 Originally Posted by Dattier Hi, The solution : Let \$\displaystyle a
I was simply pointing out that if fn(x) is convex and bounded on [a,b] with monotonically increasing (decreasing) slope bounded at a and b, THEN your proof shows that fn(x) satisfies conditions of Ascoli's theorwem. Under these conditions it is a nice proof.

Note: these are not the conditions specified in OP which remains unanswered.*

The OP should really read, if fn(x) are convex and bounded on [a,b], do they satisfy conditions for Ascoli's theorem. Yes: your theorem under underlined conditions.

*I doubt that there is such a thing as a bounded convex function with monotonically increasing (decreasing) slope on R.

Cordially.

 June 20th, 2017, 11:11 AM #14 Member   Joined: May 2017 From: France Posts: 57 Thanks: 1 Dear Zylo, It suffices to have the sequence bounded simply and not uniformly. Bye Last edited by Dattier; June 20th, 2017 at 11:19 AM.
 June 20th, 2017, 11:32 AM #15 Member   Joined: May 2017 From: France Posts: 57 Thanks: 1 For your "doubt" https://fr.wikipedia.org/wiki/Foncti...nction_convexe
June 20th, 2017, 11:57 AM   #16
Senior Member

Joined: Mar 2015
From: New Jersey

Posts: 1,175
Thanks: 90

Quote:
 Originally Posted by Dattier For your "doubt" https://fr.wikipedia.org/wiki/Foncti...nction_convexe

That simply defines a convex function (wicki translation) and its properties, and is unrelated to your incomplete proof, wherever it came from.

 June 20th, 2017, 12:01 PM #17 Member   Joined: May 2017 From: France Posts: 57 Thanks: 1 I don't have give a proof, but an explanation.

 Tags beautiful, convex, functions, result

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post Eureka Art 9 December 12th, 2015 03:10 AM Vasily Applied Math 1 June 23rd, 2012 02:11 PM Rak Real Analysis 1 December 1st, 2009 09:45 AM vikram Applied Math 0 February 4th, 2009 03:27 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top