Re: [HM] Fundamental Theorem of Algebra


Subject: Re: [HM] Fundamental Theorem of Algebra
From: John Conway (conway@math.Princeton.EDU)
Date: Sat Apr 08 2000 - 14:05:31 EDT


   The simplest proof of the "fundamental theorem" has always
been this:

   1) prove that for any real C we can find an R such that

      |f(z)| > C for |z| > R.

 {This is easy - if the polynomial is

         z^n + az^n-1 + ... + k

then if |z| > C( 1 + |1| + ... + |k|) I think we're OK.

   2) So the infimum of |f(z)| over the whole plane equals
       that over some closed disc |z| =< R.

   3) Prove or quote that the infimum of a continuous function
      over a closed and bounded set is attained. [The most
      easily-comprehended proof is by the bisection method
      of the Bolzano-Weierstrass theorem.]

   4) So we can suppose that there's a particular value Z
      of z at which |f(z)| is minimal. Shift to make Z = 0.

     We now want to prove f(0) = 0.

   5) If not, we can suppose f(0) = 1, so f looks like

      1 + az + bz^2 + ... .

   Now if only |z| is sufficiently small we have

       |bz^2 + ... | < |az|/2.

   6) Now by chose of arg(z) we can make az be negative real,

      and then if it's small enough, it's easy using 5) to
      see that | 1 + az + bz^2 + ... | < 1, contradicting
      our assumption.

   John Conway



This archive was generated by hypermail 2b28 : Sat Apr 08 2000 - 14:10:55 EDT