home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!yktnews!victor
- From: victor@watson.ibm.com (Victor Miller)
- Subject: Re: An interesting limit problem.
- Sender: news@watson.ibm.com (NNTP News Poster)
- Message-ID: <VICTOR.92Jul29153236@terse4.watson.ibm.com>
- In-Reply-To: elkies@ramanujan.harvard.edu's message of 25 Jul 92 20:18:04 EDT
- Date: Wed, 29 Jul 1992 19:32:36 GMT
- Reply-To: victor@watson.ibm.com
- Disclaimer: This posting represents the poster's views, not necessarily those of IBM
- References: <1992Jul25.212844.1@lure.latrobe.edu.au>
- <1992Jul25.201805.14172@husc3.harvard.edu>
- Nntp-Posting-Host: terse4.watson.ibm.com
- Organization: IBM, T.J. Watson Research Center
- Lines: 40
-
- >>>>> On 25 Jul 92 20:18:04 EDT, elkies@ramanujan.harvard.edu (Noam Elkies) said:
-
- Noam> In article <1992Jul25.212844.1@lure.latrobe.edu.au>
- Noam> mattm@lure.latrobe.edu.au writes:
-
- >A challenge to all mathematicians. A 100 years ago, this would probably have
- >been solved fairly simply in a natural way, but can you? I think that this
- >problem was first posed by the Russian mathematician Arnold. Hope you find this
- >problem as interesting as I did when I first solved it.
- >
- > sin(tan x) - tan(sin x)
- > lim ---------------------------------- = ???
- > x->0 arcsin(arctan x) - arctan(arcsin x)
-
- Noam> I don't know how they sould have done this 100 years ago. One direct
- Noam> approach is to expand both numerator and denominator in Taylor series
- Noam> about x=0 and compare leading terms. Since we're doing this today instead
- Noam> of 100 years ago, we can obtain the Taylor expansions instantaneously using
- Noam> a symbolic manipulation package. We find that both numerator and denominator
- Noam> are -x^7/30+O(x^9), so the limit equals 1. This also suggests that you
- Noam> *don't* want to find this limit using the L'Hospital rule, especially if
- Noam> you have to do all the derivatives by hand!
-
- This problem is rather interesting, in that it is true in the
- following general sense: let f and g be distinct elements of k[[x]]
- (the power series ring, k is an arbitray field), which have 0 constant
- term, and linear term x. Then the Laurent series
-
- f(g(x)) - g(f(x))
- -------------------------------------
- f^{-1}(g^{-1}(x)) - g^{-1}(f^{-1}(x))
-
- starts 1 + ...
- Noam> --Noam D. Elkies (elkies@zariski.harvard.edu)
- Noam> Dept. of Mathematics, Harvard University
- --
- Victor S. Miller
- Vnet and Bitnet: VICTOR at WATSON
- Internet: victor@watson.ibm.com
- IBM, TJ Watson Research Center
-