Identity theorem

From The Right Wiki
Jump to navigationJump to search

In real analysis and complex analysis, branches of mathematics, the identity theorem for analytic functions states: given functions f and g analytic on a domain D (open and connected subset of or ), if f = g on some SD, where S has an accumulation point in D, then f = g on D.[1] Thus an analytic function is completely determined by its values on a single open neighborhood in D, or even a countable subset of D (provided this contains a converging sequence together with its limit). This is not true in general for real-differentiable functions, even infinitely real-differentiable functions. In comparison, analytic functions are a much more rigid notion. Informally, one sometimes summarizes the theorem by saying analytic functions are "hard" (as opposed to, say, continuous functions which are "soft").[citation needed] The underpinning fact from which the theorem is established is the expandability of a holomorphic function into its Taylor series. The connectedness assumption on the domain D is necessary. For example, if D consists of two disjoint open sets, f can be 0 on one open set, and 1 on another, while g is 0 on one, and 2 on another.

Lemma

If two holomorphic functions f and g on a domain D agree on a set S which has an accumulation point c in D, then f=g on a disk in D centered at c. To prove this, it is enough to show that f(n)(c)=g(n)(c) for all n0, since both functions are analytic. If this is not the case, let m be the smallest nonnegative integer with f(m)(c)g(m)(c). By holomorphy, we have the following Taylor series representation in some open neighborhood U of c:

(fg)(z)=(zc)m[(fg)(m)(c)m!+(zc)(fg)(m+1)(c)(m+1)!+]=(zc)mh(z).

By continuity, h is non-zero in some small open disk B around c. But then fg0 on the punctured set B{c}. This contradicts the assumption that c is an accumulation point of {f=g}. This lemma shows that for a complex number a, the fiber f1(a) is a discrete (and therefore countable) set, unless fa.

Proof

Define the set on which f and g have the same Taylor expansion: S={zD|f(k)(z)=g(k)(z) for all k0}=k=0{zD|(f(k)g(k))(z)=0}. We'll show S is nonempty, open, and closed. Then by connectedness of D, S must be all of D, which implies f=g on S=D. By the lemma, f=g in a disk centered at c in D, they have the same Taylor series at c, so cS, S is nonempty. As f and g are holomorphic on D, wS, the Taylor series of f and g at w have non-zero radius of convergence. Therefore, the open disk Br(w) also lies in S for some r. So S is open. By holomorphy of f and g, they have holomorphic derivatives, so all f(n),g(n) are continuous. This means that {zD|(f(k)g(k))(z)=0} is closed for all k. S is an intersection of closed sets, so it's closed.

Full characterisation

Since the identity theorem is concerned with the equality of two holomorphic functions, we can simply consider the difference (which remains holomorphic) and can simply characterise when a holomorphic function is identically 0. The following result can be found in.[2]

Claim

Let G denote a non-empty, connected open subset of the complex plane. For h:G the following are equivalent.

  1. h0 on G;
  2. the set G0={zGh(z)=0} contains an accumulation point, z0;
  3. the set G=n0Gn is non-empty, where Gn:={zGh(n)(z)=0}.

Proof

The directions (1 2) and (1 3) hold trivially. For (3 1), by connectedness of G it suffices to prove that the non-empty subset, GG, is clopen (since a topological space is connected if and only if it has no proper clopen subsets). Since holomorphic functions are infinitely differentiable, i.e. hC(G), it is clear that G is closed. To show openness, consider some uG. Consider an open ball UG containing u, in which h has a convergent Taylor-series expansion centered on u. By virtue of uG, all coefficients of this series are 0, whence h0 on U. It follows that all n-th derivatives of h are 0 on U, whence UG. So each uG lies in the interior of G. Towards (2 3), fix an accumulation point z0G0. We now prove directly by induction that z0Gn for each n0. To this end let r(0,) be strictly smaller than the convergence radius of the power series expansion of h around z0, given by k0h(k)(z0)k!(zz0)k. Fix now some n0 and assume that z0Gk for all k<n. Then for zB¯r(z0){z0} manipulation of the power series expansion yields

h(n)(z0)=n!h(z)(zz0)n(zz0)n!k=n+1h(k)(z0)k!(zz0)k(n+1)=:R(z). (1)

Note that, since r is smaller than radius of the power series, one can readily derive that the power series R() is continuous and thus bounded on B¯r(z0). Now, since z0 is an accumulation point in G0, there is a sequence of points (z(i))iG0Br(z0){z0} convergent to z0. Since h0 on G0 and since each z(i)G0Br(z0){z0}, the expression in (1) yields

h(n)(z0)=n!h(z(i))(z(i)z0)n(z(i)z0)R(z(i))=0(z(i)z0)i0R(z(i)). (2)

By the boundedness of R() on B¯r(z0), it follows that h(n)(z0)=0, whence z0Gn. Via induction the claim holds. Q.E.D.

See also

References

  1. For real functions, see Krantz, Steven G.; Parks, Harold R. (2002). A Primer of Real Analytic Functions (Second ed.). Boston: Birkhäuser. Corollary 1.2.7. ISBN 0-8176-4264-1.
  2. Guido Walz, ed. (2017). Lexikon der Mathematik (in Deutsch). Vol. 2. Mannheim: Springer Spektrum Verlag. p. 476. ISBN 978-3-662-53503-5.
  • Ablowitz, Mark J.; Fokas A. S. (1997). Complex variables: Introduction and applications. Cambridge, UK: Cambridge University Press. p. 122. ISBN 0-521-48058-2.