On the coefficient formula for de Branges–Rovnyak norms
Thomas Ransford · 2026
Read On the coefficient formula for de Branges–Rovnyak norms by Thomas Ransford (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-30114.txt.
Introduction
Thomas Ransford extends a coefficient formula for de Branges-Rovnyak norms. The formula expresses a norm through Taylor coefficients of a function and a related one. It was known for functions holomorphic near the closed disk, but can fail for others in the space. Ransford extends it to square-integrable functions with finite right side, if the related one is square-integrable or rational. Under weaker integrability, it still holds given an extra weighted finiteness condition. He gives a limit form valid for all non-extreme points and functions there.
Catalog record
| CREATOR | Thomas Ransford |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 49408 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.30114
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the coefficient formula for de Branges–Rovnyak norms Thanks: Research supported by NSERC Discovery Grant RGPIN-2026-04565 Thomas Ransford Abstract . Let ℋ ( b ) {\mathcal{H}}(b) be the de Branges–Rovnyak space associated to a non-extreme point b b of the unit ball of H ∞ H^{\infty} , and let ϕ = b / a \phi=b/a , where a a is the Pythagorean mate of b b . It is known that, if f f is a function holomorphic on a neighbourhood of the closed unit disk, then it belongs to ℋ ( b ) {\mathcal{H}}(b) , and its norm in ℋ ( b ) {\mathcal{H}}(b) can be expressed in terms of the Taylor coefficients of f f and ϕ \phi via the formula ‖ f ‖ ℋ ( b ) 2 = ∑ m ≥ 0 | f ^ ( m ) | 2 + ∑ m ≥ 0 | ∑ n ≥ 0 ϕ ^ ( n ) ¯ f ^ ( m + n ) | 2 . \|f\|_{{\mathcal{H}}(b)}^{2}=\sum_{m\geq 0}|\widehat{f}(m)|^{2}+\sum_{m\geq 0}\Bigl|\sum_{n\geq 0}\overline{\widehat{\phi}(n)}\widehat{f}(m+n)\Bigr|^{2}. However, the formula can break down for some other f ∈ ℋ ( b ) f\in{\mathcal{H}}(b) . In this article we extend the scope of the formula to all f ∈ H 2 f\in H^{2} for which the right-hand side is finite, provided that either ϕ ∈ H 2 \phi\in H^{2} or ϕ \phi is rational. If merely ϕ ∈ H p \phi\in H^{p} for some p ∈ ( 0 , 2 ] p\in(0,2] , then the formula still holds provided that, in addition, ∑ m ≥ 0 m 2 / p − 1 | f ^ ( m ) | 2 < ∞ \sum_{m\geq 0}m^{2/p-1}|\widehat{f}(m)|^{2}<\infty . We also establish a limit-form of the formula that is valid for all non-extreme b b and all f ∈ ℋ ( b ) f\in{\mathcal{H}}(b) . 1991 Mathematics Subject Classification Primary 47B32, Secondary 47B35 1. Introduction The spaces now called de Branges–Rovnyak spaces were introduced by de Branges and Rovnyak in the appendix of [ 5 ] and further studied in [ 6 ] . They were later popularized by the book of Sarason [ 18 ] . It is now known that these spaces have a variety of connections with other topics in complex analysis and operator theory, and they have attracted the interest of many authors. When working with de Branges–Rovnyak spaces, it is very helpful to have a formula for the norm (and, via polarization, for the inner product) of functions in these spaces, expressed in terms of their Taylor coefficients. There does indeed exist such a formula, namely, (1) \|f\|_{{\mathcal{H}}(b)}^{2}\text{~``=''~}\sum_{m\geq 0}|\widehat{f}(m)|^{2}+\sum_{m\geq 0}\Bigl|\sum_{n\geq 0}\overline{\widehat{\phi}(n)}\widehat{f}(m+n)\Bigr|^{2}. Here {\mathcal{H}}(b) is the de Branges–Rovnyak space associated to a non-extreme point b of the unit ball of H^{\infty} , and \phi=b/a , where a is the Pythagorean mate of b (all these terms will be explained in § 2 below). Also \widehat{f}(n),\widehat{\phi}(n) are the n -th coefficients of the Taylor expansions around 0 of f,\phi respectively. What is noteworthy about the formula ( 1 ) is that it is only sometimes true—hence the inverted commas around the equals sign. It is always true if f is a polynomial. Indeed, the special case when f(z)=z^{k} , namely \|z^{k}\|_{{\mathcal{H}}(b)}=1+\sum_{n=0}^{k}|\widehat{\phi}(n)|^{2}, is well known and used in many places in the literature. More generally, equality holds in ( 1 ) if f is holomorphic in a neighbourhood of the closed unit disk; this is implicit in [ 20 ] , and proved explicitly in [ 3 ] . As polynomials are dense in {\mathcal{H}}(b) , one might be tempted to infer that same formula holds for all f\in{\mathcal{H}}(b) , but in fact this turns out to be false in general. Counterexamples can be found, for example, in [ 8 ] and [ 16 ] , and there are more in this article. In view of the potential applications, it is of great interest to widen the scope of validity of the formula to include as many functions f as possible. This is the goal of the present paper. Here is a summary of our main results: • If \phi belongs to the Hardy space H^{2} , then equality holds in ( 1 ) for all f\in{\mathcal{H}}(b) (Theorem 3.1 ). On the other hand, this is no longer necessarily the case if merely \phi\in\cap_{p<2}H^{p} (Corollary 6.2 ). • If \phi belongs to H^{p} for some p\in(0,2] , then equality holds in ( 1 ) for all f\in H^{2} for which the right-hand side of ( 1 ) is finite and such that, in addition, \sum_{m\geq 0}m^{2/p-1}|\widehat{f}(m)|^{2}<\infty (Theorem 4.1 ). • If \phi is rational then equality holds in ( 1 ) for all f\in H^{2} such that the right-hand side of ( 1 ) is finite. If, in addition, all the poles of \phi on the unit circle {\mathbb{T}} are simple, then equality holds in ( 1 ) for all f\in{\mathcal{H}}(b) . On the other hand, if \phi has a multiple pole on {\mathbb{T}} , then ( 1 ) breaks down for at least one f\in{\mathcal{H}}(b) (Theorem 5.1 ). • There is a limit form of the formula ( 1 ) that is valid for all all Pythagorean pairs (b,a) and all f\in{\mathcal{H}}(b) (Theorem 7.1 ). 2. Background on de Branges–Rovnyak spaces In this section we review de Branges–Rovnyak spaces and some related notions that will be needed in the sequel. The treatment is deliberately brief. The interested reader may find much more information in the books of Sarason [ 18 ] and Fricain–Mashreghi [ 10 ] . 2.1. De Branges–Rovnyak spaces Let b\in H^{\infty} with \|b\|_{H^{\infty}}\leq 1 . Then the map (z,w)\mapsto(1-\overline{b(w)}b(z))/(1-\overline{w}z) is positive-definite on {\mathbb{D}}\times{\mathbb{D}} , so it is the reproducing kernel of a unique Hilbert function space on {\mathbb{D}} . This space is called the de Branges–Rovnyak space associated to b , and is denoted by {\mathcal{H}}(b) . It is always contractively contained in the Hardy space H^{2} . The theory of de Branges–Rovnyak spaces is governed by a fundamental dichotomy, namely whether or not b is an extreme point of the unit ball of H^{\infty} . For example, if b is non-extreme, then {\mathcal{H}}(b) contains the polynomials and they are dense in {\mathcal{H}}(b) . On the other hand, if b is extreme, then the polynomials in {\mathcal{H}}(b) only span a finite-dimensional subspace. In this article, we shall be concerned exclusively with the case where b is non-extreme. 2.2. Pythagorean pairs We say that (b,a) is a Pythagorean pair if • b,a\in H^{\infty} ; • |b|^{2}+|a|^{2}=1 a.e. on {\mathbb{T}} ; • a is outer and a(0)>0 . The function a is completely determined by b , and is called its Pythagorean mate . A function b occurs as the first element of a Pythagorean pair iff it is a non-extreme point of the unit ball of H^{\infty} . Thus the map (b,a)\mapsto b is a bijection between the set of Pythagorean pairs and the set of non-extreme points of the unit ball of H^{\infty} . Given a Pythagorean pair (b,a) , we write T_{\overline{b}} and T_{\overline{a}} for the bounded Toeplitz operators on H^{2} with symbols \overline{b} and \overline{a} respectively. The following