home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!pavo.csi.cam.ac.uk!gjm11
- From: gjm11@cus.cam.ac.uk (G.J. McCaughan)
- Newsgroups: sci.math
- Subject: Re: Do you know ?
- Message-ID: <1992Oct16.164029.17849@infodev.cam.ac.uk>
- Date: 16 Oct 92 16:40:29 GMT
- References: <1992Oct16.050019.25103@noose.ecn.purdue.edu>
- Sender: news@infodev.cam.ac.uk (USENET news)
- Organization: U of Cambridge, England
- Lines: 21
- Nntp-Posting-Host: bootes.cus.cam.ac.uk
-
- In article <1992Oct16.050019.25103@noose.ecn.purdue.edu> kavuri@lips.ecn.purdue.edu (Surya N Kavuri ) writes:
-
- > Can you give me a function on an interval [a b] which is
- > not differentiable anywhere on that interval ?
-
- I presume you want it to be continuous too.
-
- For k=1,2,3,... let f_k be the function whiose graph starts at (a,0) and
- then zig-zags up and down with slope alternately +1 and -1, changing direction
- every 2^(-k).(a-b). So the graph of f_k has 2^(k-1) pointy tops, and its
- maximum value is 2(-k).
-
- Let f be the sum over all k of f_k. Then it's very easy to show that f is
- continuous; but also f is not differentiable anywhere; to show f is not
- differentiable at x, take successively nearer and nearer points of the form
- a+t(a-b)/2^k (there are points of this form arbitrarily near x) and look at
- the gradients of the chords joining x to them, which go haywire.
-
- --
- Gareth McCaughan Dept. of Pure Mathematics & Mathematical Statistics,
- gjm11@cus.cam.ac.uk Cambridge University, England. [Research student]
-