🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: alt.uu.math.misc
Article #25

Bound for mixed partial derivatives

#25
From: Enrico Talinucci
Date: Fri, 28 Jan 2000 00:00
43 lines
2574 bytes
Hi everybody!

I'm in trouble with the following problem.

f: R^2 -> R is an infinitely differentiable function. I know an upper
bound for the modulus of its pure partial derivatives, i.e.
|d^n f/dx^n| and |d^n f/dy^n|, for every (x,y) \in R^2. I need to find a
bound on its mixed p.d., d^(k+h) f /(dx^k dy^h).


Now I'll detail the question.
The data is:

   |d^k f/dx^k|  < A_x ^k ||(x,y)||
   |d^k f/dy^k|  < A_y ^k ||(x,y)||

for every (x,y) and every k, where A_x, A_y are known constants
and ||..|| is a norm on R^2, let's say the Euclidean norm. The
inequalities holds for k=0 too.
The target is finding a bound B(k, h, A_x, A_y) so that:

   |d^(k+h) f /(dx^k dy^h)| < B(k, h, A_x, A_y) ||(x,y)||

for every (x,y), k, h.


An hypotesis could be:

  B = A_x^k  A_y^h

I couldn't prove it neither find a counterexample. It's just to give an
idea of what I'm looking for.


Does such a bound exist? If so, can you help me to find it? If not,
can you give me a counterexample, i.e. a function who satisfies the
given bounds, but for which at least one m.p.d. is unbounded with
respect to ||(x,y)||?


Thanks in advance for your help.

Enrico Talinucci
Technical Details
Message-ID: <86qqcr$hpo$1@nslave2.tin.it>
Path: arc1-ca.pugleaf.net!unknown!not-for-mail

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 109d8a,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid109d8a,public
X-Google-Thread: f5c59,d6aa6ea11a63fbeb,start
X-Google-Attributes: gidf5c59,public
X-Google-Thread: 10d67e,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid10d67e,public
X-Google-Thread: 11339a,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid11339a,public
X-Google-Thread: 1077c2,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid1077c2,public
X-Google-Thread: 10c033,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid10c033,public
X-Google-Thread: 107079,d6aa6ea11a63fbeb,start
X-Google-Attributes: gid107079,public
From: "Enrico Talinucci" <talinucci@tin.it>
Subject: Bound for mixed partial derivatives
Date: 2000/01/28
Message-ID: <86qqcr$hpo$1@nslave2.tin.it>
X-Deja-AN: 578595901
Sender: Daniel Grayson, dan@math.uiuc.edu, moderator for sci.math.research
X-Priority: 3
X-Coding-System: undecided-unix
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211
X-Complaints-To: abuse@uiuc.edu
X-Trace: vixen.cso.uiuc.edu 949030198 128.174.111.4 (Thu, 27 Jan 2000 21:29:58 CST)
Organization: University of Illinois at Urbana-Champaign
X-MSMail-Priority: Normal
NNTP-Posting-Date: Thu, 27 Jan 2000 21:29:58 CST
Newsgroups: alt.uu.math.misc,aus.mathematics,sci.math,sci.math.num-analysis,sci.math.symbolic,tnn.math,umn.math.dept