Abel's theorem
≈ 7 min read
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
In mathematics, Abel's theorem for power series relates a limit of a power series to the sum of its coefficients. It is named after Norwegian mathematician Niels Henrik Abel, who proved it in 1826.
In mathematics, Abel's theorem for power series relates a limit of a power series to the sum of its coefficients. It is named after Norwegian mathematician Niels Henrik Abel, who proved it in 1826.
Theorem
Let the Taylor series G (x) = \sum_{k=0}^\infty a_k x^k be a power series with real coefficients a_k . Suppose that the series \sum_{k=0}^\infty a_k converges. Then G(x) is continuous from the left at x = 1, that is, \lim_{x\to 1^-} G(x) = \sum_{k=0}^\infty a_k. The same theorem holds for complex power series G(z) = \sum_{k=0}^\infty a_k z^k, provided that z \to 1 entirely within a single Stolz sector, that is, a region of the open unit disk where |1-z| \leq M(1-|z|) for some fixed finite M > 1 .
Without this restriction, the limit may fail to exist: for example, the power series \sum_{n>0} \frac{z^{3^n}-z^{2\cdot 3^n}} n converges to 0 at z = 1, but is unbounded near any other point of the form e^{\pi i/3^n}, so the value at z = 1 is not the limit as z tends to 1 in the whole open disk. The convergence of \sum_{k=0}^\infty a_k implies that the radius of convergence of the power series \sum_{k=0}^{\infty} a_k z^k is at least 1, ensuring convergence for |z| .
Also, by the uniform limit theorem, G(z) is continuous on the real closed interval [0, t] for t by virtue of the uniform convergence of the series on compact subsets of the disk of convergence (by the Weierstrass M-test). Abel's theorem allows us to say more, namely that the restriction of G(z) to [0, 1] is continuous.
The Stolz sector |1-z|\leq M(1-|z|) has explicit equation y^2 = -\frac{M^4 (x^2 - 1) - 2 M^2 ((x - 1) x + 1) + 2 \sqrt{M^4 (-2 M^2 (x - 1) + 2 x - 1)} + (x - 1)^2}{(M^2 - 1)^2} and is plotted on the right for various values. The left end of the sector is x = \frac{1-M}{1+M} , and the right end is x=1 .
On the right end, it becomes a cone with angle 2\theta where \cos\theta = \frac{1}{M} .
Remarks
As an immediate consequence of this theorem, if z is any nonzero complex number for which the series \sum_{k=0}^\infty a_k z^k converges, then it follows that \lim_{t\to 1^{-}} G(tz) = \sum_{k=0}^\infty a_kz^k in which the limit is taken from below. The theorem can also be generalized to account for sums which diverge to infinity; see below.
If \sum_{k=0}^\infty a_k = \infty then \lim_{x \to 1^{-}} G(x) \to \infty. However, if the series is only known to be divergent, but for reasons other than diverging to infinity, then the claim of the theorem may fail: take, for example, the power series for \frac{1}{1+z}. At z = 1 the series is equal to 1 - 1 + 1 - 1 + \cdots, but \tfrac{1}{1+1} = \tfrac{1}{2}.
We also remark the theorem holds for radii of convergence other than R = 1 : let G(x) = \sum_{k=0}^\infty a_kx^k be a power series with radius of convergence R > 0, and suppose the series converges at x = R. Then G(x) is continuous from the left at x = R, that is, \lim_{x\to R^-}G(x) = G(R).
Applications
Abel's theorem allows us to evaluate many series in closed form if we can first prove that they converge. For example, consider the sequence a_k = \frac{(-1)^k}{k+1}.
The sum \sum_{k=0}^\infty a_k converges by the alternating series test, so we can apply Abel's theorem to the corresponding generating function G_a(z) . We obtain G_a(z) = \frac{\ln(1+z)}{z}, \qquad 0 by integrating the uniformly convergent geometric power series term by term on [-z, 0] ; thus the series \sum_{k=0}^\infty \frac{(-1)^k}{k+1} converges to \lim \limits_{z \to 1^-} G_a(z) = \ln 2 by Abel's theorem. Similarly, \sum_{k=0}^\infty \frac{(-1)^k}{2k+1} converges to \arctan 1 = \tfrac{\pi}{4}.
G_a(z) is called the generating function of the sequence a. Abel's theorem is frequently useful in dealing with generating functions of real-valued and non-negative sequences, such as probability-generating functions. In particular, it is useful in the theory of Galton–Watson processes.
Outline of proof
Source: Let S = \sum_{k=0}^\infty a_k and s_n= \left({\sum_{k=0}^n a_k}\right) - S\!. Then substituting a_k=s_k-s_{k-1} and performing a simple manipulation of the series (summation by parts) results in G_a(z) = S + (1-z)\sum_{k=0}^{\infty} s_k z^k. Given \varepsilon > 0, pick n large enough so that |s_k| for all k \geq n. We have \left|(1-z)\sum_{k=n}^\infty s_kz^k \right| \leq \varepsilon |1-z|\sum_{k=n}^\infty |z|^k = \varepsilon|1-z|\frac when z lies within the given Stolz angle.
Whenever z is sufficiently close to 1 we have \left|(1-z)\sum_{k=0}^{n-1} s_kz^k \right| so that \left|G_a(z) - S\right| when z is both sufficiently close to 1 and within the Stolz angle. To prove the case for \sum_{k=0}^\infty a_k = +\infty , let x \in (0,1) and s_n = \sum_{k=0}^n a_k . Since (s_k)_k diverges to +\infty , we can find, for any M \in \mathbb{R}_+ , a N \in \N such that s_k > 3M for all k \geq N . We write, for n > N : \sum_{k=0}^n a_k x^k = \left[{(1-x) \sum_{k=0}^{N-1} s_k x^k}\right] + \left\{{ (1-x) \left[{ \sum_{k=N}^{n-1} s_k x^k }\right] + s_n x^n }\right\} Since the first term on the right hand side vanishes as x \to 1 , we can find an x_1 \in (0, 1) such that it exceeds -M whenever x_1 .
The second term may be estimated by: (1-x) \left[{ \sum_{k=N}^{n-1} s_k x^k }\right] + s_n x^n > 3M \left\{{ (1-x) \left[{\sum_{k=N}^{n-1} x^k}\right] + x^n}\right\} = 3M x^N. Hence, if we let x_2 = (2/3)^{1/N} , then for x_2 this exceeds 2M . Combining, we get, for any n>N and \max(x_1, x_2) : \sum_{k=0}^n a_k x^k > M. This establishes: \lim_{x \to 1^{-}} \liminf_{n \to \infty} \sum_{k=0}^n a_k x^k = +\infty In the absence of additional assumptions, the series \sum_{k=0}^{\infty} a_k x^k might not converge when |x| , hence the use of the limit inferior.
Related concepts
Converses to a theorem like Abel's are called Tauberian theorems: There is no exact converse, but results conditional on some hypothesis. The field of divergent series, and their summation methods, contains many theorems of abelian type and of tauberian type.
Generalizations
In the real case, the functions x^k may be replaced by bounded non-negative functions u_k(x) monotonically decreasing with k (i.e., u_k(x) \ge u_{k+1}(x) ) such that \lim_{x \to 1^{-}} u_k(x) = 1 for all k . This follows from Abel's uniform convergence test and applying the uniform limit theorem.
Further reading
- Ahlfors called it Abel's limit theorem.
External links
(a more general look at Abelian theorems of this type)
Text derived from https://en.wikipedia.org/w/index.php?oldid=1370158481 under CC BY-SA 4.0; modifications have been made.