home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!swrinde!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!rutgers!dziuxsolim.rutgers.edu!no.rutgers.edu!bumby
- From: bumby@no.rutgers.edu (Richard Bumby)
- Newsgroups: sci.math
- Subject: Re: Continued Fractions
- Keywords: Help, Continued Fractions
- Message-ID: <Aug.20.17.18.51.1992.1225@no.rutgers.edu>
- Date: 20 Aug 92 21:18:52 GMT
- References: <1992Aug20.165623.11309@eliot.uucp>
- Distribution: na
- Organization: Rutgers Univ., New Brunswick, N.J.
- Lines: 76
-
- andyc@eliot.uucp (Andy Collins) writes:
-
- >I am looking for information on continued fractions. I would
- >hope that there are books on the subject out there, but have
- >not had the time to do a full-out search. I would appreciate
- >any information.
-
- >Thanks
-
- >andy
-
- >PS (not a trademark)
- >Where are continued fractions used?
- >Who has done the most work on these things?
-
- >If you can could you post replies, my mail will be straigtened out
- >by next week.
-
- Continued fractions are the most exciting objects in all of
- mathematics. I would not be surprised to find love poetry inspired by
- continued fractions! As for references ...
-
- Perron's "Kettenbrueche" (1929) remains the standard reference. It
- has been reprinted by Chelsea, but appears to be available only in the
- original German. This book covers both the arithmetic and analytic
- theory.
-
- The chief use of continued fractions arises from the fact that they
- describe a procedure for efficient generation of good rational
- approximations. A good treatment of the arithmetic theory from this
- point of view can be found in Cassels, "An Introduction to Diophantine
- Approximation", Cambridge, 1957.
-
- Any reasonable thorough textbook on elementary number theory would
- have an introduction to the arithmetic theory of continued fractions.
- The analytic theory still seems to be treated somewhat as an advanced
- subject.
-
- The continued fraction representation of a real number seems to be a
- very rigid object. Whatever property you choose as the basis of your
- study, you seem to be led to the same construction. The question of
- approximation by rational numbers have analogs concerning simultaneous
- approximation of several real numbers by rationals with the same
- denominator. It is tempting to think that there should be something
- like a continued fraction that could be used to organize work on this
- subject. No comparable theory has been found, although some
- constructions have been proposed that have a large number of desirable
- properties. A very thorough discussion of this can be found in
- Brentjes, "Multidimensional Continued Fraction Algorithms",
- Mathematical Centre Tract 145 (Amsterdam, 1981).
-
- The subject is important enough to have its own section heading in the
- Encyclopedic Dictionary of Mathematics, though the 3 pages allocated
- do not give enough space to describe the subject as it relates to
- current research. The fact that the continued fraction expansion of
- every quadratic irrational is eventually periodic (published by Lagrange
- in 1770) continues to be the star of the show. Combining this with
- the result that says that the convergents of the continued fraction
- give _all_ good rational approximations (in a sense which I will not
- make precise here), shows that these periods allow the fundamental
- units of real quadratic fields to be computed.
-
- The relation between the partial quotients in the continued fraction
- of a number and the approximation properties of the number allow
- examples to be constructed of numbers having prescribed approximation
- properties. I find this to be extremely important, even though it is
- obtained fairly early in the development of the theory. Its
- importance is somewhat subtle since it depends on an interpretation of
- theorems which is not easily included in their statement. In a sense,
- the pattern of the sequence of partial quotients includes more
- significant information about the number than, for example, its string
- of decimal digits.
- --
- R. T. Bumby ** Rutgers Math || Amer. Math. Monthly Problems Editor
- bumby@math.rutgers.edu || P.O. Box 10971 New Brunswick, NJ08906-0971
- bumby@dimacs.rutgers.edu || Phone: [USA] 908 932 0277 * FAX 908 932 5530
-