home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!spool.mu.edu!sdd.hp.com!cs.utexas.edu!zaphod.mps.ohio-state.edu!not-for-mail
- From: edgar@function.mps.ohio-state.edu (Gerald Edgar)
- Newsgroups: sci.math
- Subject: Re: closed forms
- Date: 23 Nov 1992 07:56:50 -0500
- Organization: The Ohio State University, Dept. of Math.
- Lines: 36
- Message-ID: <1eqkeiINN9p5@function.mps.ohio-state.edu>
- References: <92324.223138YUKQC@CUNYVM.BITNET> <1992Nov22.100111.42575@urz.unibas.ch>
- NNTP-Posting-Host: function.mps.ohio-state.edu
-
- In article <1992Nov22.100111.42575@urz.unibas.ch> kullmann@urz.unibas.ch writes:
- >Another question that could belong here: How does one proof that there is
- >no closed form expression of \int(exp(-x**2))?
-
- This is a result of Liouville. More generally, he proved:
-
- THEOREM: Let g and y be algebraic functions of x, with g nonconstant.
- If integral exp(g) y dx is an elementary function of x, then it has
- the form exp(g) w + C, where w is a rational function of x, y, and g.
-
- A function is "elementary" if it is obtained from algebraic operations,
- trig and inverse trig functions, exponential and logarithmic functions,
- or combinations of these (including composition and inverse function).
- (There is a more technical definition...)
-
- Examples of the application of the theorem are:
- integral exp(-x^2) dx is not elementary
- integral exp(x)/x dx is not elementary
- integral dx/log x is not elementary.
-
- A book on this classical theory is: _Integration in Finite Terms_
- by J. F. Ritt (1948). The background required is complex analysis.
- There are also some more recent books, which take a more algebraic
- approach, but I do not have those references handy.
-
-
- Note. Derivation of the fact that integral exp(-x^2) dx = exp(-x^2) w + C
- for rational function w is impossible: First, w must satisfy the
- differential equation w' - 2x = 1. Deduce that w has no (finite) poles, so
- w is a polynomial. Obtain a contradiction by comparing the degrees of
- polynomials.
- --
- Gerald A. Edgar Internet: edgar@mps.ohio-state.edu
- Department of Mathematics Bitnet: EDGAR@OHSTPY
- The Ohio State University telephone: 614-292-0395 (Office)
- Columbus, OH 43210 -292-4975 (Math. Dept.) -292-1479 (Dept. Fax)
-