home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!utcsri!eecg.toronto.edu!leemike
- Newsgroups: sci.math
- From: leemike@eecg.toronto.edu (Michael Lee)
- Subject: Re: need proof: (1 + 1/n)^n ==> e
- Message-ID: <1992Dec16.023520.11420@jarvis.csri.toronto.edu>
- Organization: CSRI, University of Toronto
- References: <1glr2qINN2vg@usenet.INS.CWRU.Edu> <1glt0lINNquc@news.aero.org>
- Date: 16 Dec 92 07:35:20 GMT
- Lines: 52
-
- In article <1glt0lINNquc@news.aero.org> doner@Aero.org (John Doner) writes:
- >In article <1glr2qINN2vg@usenet.INS.CWRU.Edu>, fau@po.CWRU.Edu (Francis A. Uy) writes:
- >|>
- >|> I can't believe I forgot the proof of this one.
- >|> Subject line says it all.
- >
- >First tell us what you want to use for a definition of e.
-
- First show that (1 + 1/n)^n <= e <= (1 + 1/n)^(n+1).
- b
- |\
- Let Integ[a, b, f(t) dt] = | f(t) dt
- \|
- a
- n is a positive Integer.
-
- log(1 + 1/n) = Integ[1, 1+1/n, dt/t]
-
- Since 1 > 1/t in the interval of t,
- Integ[1, 1+1/n, dt/t] <= Integ[1, 1+1/n, dt],
- Integ[1, 1+1/n, dt] = 1/n.
-
- Since 1/(1+1/n) => 1/t in the interval of t,
- Integ[1, 1+1/n, dt/t] => Integ[1, 1+1/n, dt/(1+1/n)],
- Integ[1, 1+1/n, dt/(1+1/n)] = 1/(1+n).
-
- Therefore,
- 1/(n+1) <= log(1+1/n) <= 1/n.
-
- From the right inequality,
- 1 + 1/n <= e^(1/n);
- thus,
- (1 + 1/n)^n <= e.
- From the left inequality,
- e^(1/(1+n)) <= 1 + 1/n;
- thus,
- e <= (1 + 1/n)^(n+1).
-
- Therefore,
- (1+1/n)^n <= e <= (1+1/n)^(n+1)
-
- Divide the right inequality by 1+1/n,
- e/(1+1/n) <= (1+1/n)^n.
- Combine the above with the left inequality
- e/(1+1/n) <= (1+1/n)^n <= e.
- As n approaches infinity, e/(1+1/n) -> e.
-
- Therefore, using the pinching theorem of limits,
- (1+1/n)^n -> e.
-
- regards,
- Michael Lee
-