Bounded Toeplitz Products on the Hardy Space
Ryan O’Loughlin · 2025
Read Bounded Toeplitz Products on the Hardy Space by Ryan O’Loughlin (2025) 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-2501-14069.txt.
Introduction
Ryan O'Loughlin settles a 2025 question about when the product of two Toeplitz operators stays bounded on the Hardy space. Toeplitz operators are a central class in analysis, acting on functions on the unit circle. A single such operator is bounded exactly when its symbol is bounded, a clean rule known for long.
For a pair, no rule in terms of the symbols was known that matched boundedness of the product exactly. O'Loughlin shows that, under stated assumptions, the product is bounded exactly when the product of the two symbols is bounded. That turns a two operator question into a check on one product function. The result closes a gap in the theory of these operators and gives later work a firm test to use.
Catalog record
| CREATOR | Ryan O’Loughlin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-14069
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bounded Toeplitz Products on the Hardy Space Ryan O’Loughlin Abstract A Toeplitz operator on the Hardy space of the unit circle is bounded if and only if its symbol is bounded. For two Toeplitz operators, there are no known function-theoretic conditions for their symbols, which are equivalent to the product of the Toeplitz operators being bounded. In this paper, we provide a solution to this problem, by showing under certain assumptions that the product of two Toeplitz operators is bounded if and only if the product of their symbols is bounded. Keywords: Toeplitz operator, Hardy space, Banach space operators. MSC: 47B35, 30H10, 47B48. 1 Introduction The boundedness of Toeplitz operators was characterized by Otto Toeplitz in 1912 [ 26 , 27 ] with his elegant characterization, showing that a Toeplitz operator is bounded if and only if its symbol is bounded (see Section 1.1 in [ 19 ] for a detailed history of the topic). This led to a rich active area of research at the intersection between Operator Theory and Complex Analysis. Despite more than a century of research into Toeplitz operators, there is no function-theoretic characterisation for the product of Toeplitz operators being bounded. The purpose of this paper is to provide such a characterisation, by showing under certain criteria, the product of Toeplitz operators is bounded if and only if the product of their symbols is bounded. In 1994, Sarason posed a more specialized version of the problem in the problem book [ 11 ] . Question 1.1 (Problem 7.9 [ 11 ] ) . Characterize the pairs of outer functions u,v in H^{2} of the unit disk such that the operator T_{u}T_{\overline{v}} is bounded on H^{2} . Over the past three decades there have been many contributions to this open question in the context of the Hardy, Bergman and Fock spaces, for example, [ 8 , 28 , 17 , 25 , 23 , 2 , 5 , 4 , 24 ] as well as many others. Although Toeplitz operators have a well-established theory, questions on Toeplitz products remain particularly challenging (see, for example, the zero Toeplitz product problem [ 3 ] ). Characterising the boundedness of products of Toeplitz operators is equivalent to a problem in Harmonic Analysis, which asks when the two-weighted Riesz projection is bounded (see Section 5 for a detailed explanation of this). Since the setup and techniques we pursue lend themselves better to function theory, we approach the problem from the perspective of Toeplitz operators and express our results in terms of the two-weighted Riesz projection in Section 5 . The main new result is Theorem 4.2 , which states that when the symbols of two Toeplitz operators are an admissible Toeplitz pair (defined as Definition 4.1 ) the product of the Toeplitz operators is bounded if and only if the product of their symbols is bounded. In Section 2 we provide background on Hardy spaces and Toeplitz operators. Section 3 develops new theory on the poles of functions in the Hardy space, which is crucial for the proof of Theorem 4.2 . Section 4 presents the proof of Theorem 4.2 . Section 5 expresses new results in terms of the two-weighted Riesz projection and proves a conjecture of Sarason on Question 1.1 when the symbols of the operators are an admissible Toeplitz pair. Section 5 also shows the product of Toeplitz operators is bounded if and only if a certain weighted weighted Lebesgue measure is a Carleson measure for the range space of a Toeplitz operator. 2 Preliminaries 2.1 Hardy Space Theory In order to prove Theorem 4.2 one needs an understanding of the Hardy spaces, denoted H^{p}(\mathbb{T}) , for 0<p\leqslant\infty . Due the H^{p}(\mathbb{T}) spaces for p<1 being less well known, we include a brief discussion of these spaces. An excellent source for such spaces (in which all the following results can be found) is [ 6 ] . For p\in(0,\infty) , the Hardy space H^{p}(\mathbb{T}) is the set of analytic functions f on the open unit disk, \mathbb{D}:=\{z\in\mathbb{C}:|z|<1\} , for which the quantity \|f\|_{p}:=\sup_{0<r<1}\left\{\int_{0}^{2\pi}\left|f\left(re^{i\theta}\right)\right|^{p}\frac{d\theta}{2\pi}\right\}^{1/p} is finite. For p\in[1,\infty) , the quantity \|f\|_{p} defines a norm on H^{p}(\mathbb{T}) which makes it a Banach space; while for p\in(0,1) , the metric \rho(f,g):=\|f-g\|_{p}^{p} makes H^{p}(\mathbb{T}) a a complete topological vector space with topology given by a translation invariant metric (Theorem 3.2.7 in [ 6 ] ). For f\in H^{p}(\mathbb{T}) where 0<p<\infty , Theorem 3.2.3 in [ 6 ] says that the radial limit f(\zeta):=\lim_{r\rightarrow 1^{-}}f(r\zeta) exists for almost every \zeta\in\mathbb{T}:=\{z\in\mathbb{C}:|z|=1\} and furthermore for \mathrm{d}m:=\mathrm{d}\theta/2\pi , the normalised Lebesgue measure on the unit circle \mathbb{T} , \|f\|_{p}:=\sup_{0<r<1}\left\{\int_{\mathbb{T}}|f(r\zeta)|^{p}dm(\zeta)\right\}^{1/p}=\left\{\int_{\mathbb{T}}|f(\zeta)|^{p}dm(\zeta)\right\}^{1/p}. (2.1) With ( 2.1 ), we will freely identify a function f in the H^{p}(\mathbb{T}) space with its radial boundary values on \mathbb{T} , thus viewing the H^{p}(\mathbb{T}) space as lying inside the space L^{p}(\mathbb{T}) . H^{2}(\mathbb{T}) is a RKHS with reproducing kernel at x\in\mathbb{D} given by \tilde{k_{x}}(z):=\frac{1}{1-\overline{x}z} meaning that for f\in H^{2}(\mathbb{T}) , we have f(x)=\left\langle f,\tilde{k_{x}}\right\rangle_{H^{2}} We say a function h\in H^{p}(\mathbb{T}) is inner if |h|=1 a.e. on \mathbb{T} . An outer function is a holomorphic function on \mathbb{D} , G such that G(z)=c\exp\left(\int_{\mathbb{T}}\frac{\zeta+z}{\zeta-z}\log\left(\varphi\left(\zeta\right)\right)\mathrm{d}m(\zeta)\right) for some c\in\mathbb{C} with |c|=1 , and some function \varphi on \mathbb{T} such that \log(\varphi) is integrable on \mathbb{T} . A classical result (which can be found as Theorem 3.2.4 in [ 6 ] ) states that for 0<p<\infty every f\in H^{p}(\mathbb{T}) can be factorised so that f=f^{i}f^{0} , where f^{i} is an inner function and f^{o} is an outer function. For a function f\in L^{p}(\mathbb{T}) , where 1\leqslant p\leqslant\infty , define the Riesz projection P by (Pf)(z):=\sum_{n=0}^{\infty}\widehat{f}(n)z^{n}=\int_{\mathbb{T}}\frac{f(\zeta)}{1-\overline{\zeta}z}dm(\zeta),\quad z\in\mathbb{D}, where \widehat{f}(n):=\int_{\mathbb{T}}f(\zeta)\overline{\zeta}^{n}dm(\zeta) . Observe that as |\widehat{f}(n)|=O(1) for all f\in L^{1}(\mathbb{T}) , P(f) defines an analytic function in \mathbb{D} . Definition 2.1 . H(\mathbb{T}) is the space of holomorphic functions on \mathbb{D} that have a.e. defined radial boundary values on \mathbb{T} . We equip this space with the topology of uniform convergence on compact subsets, i.e., f_{n}\rightarrow f_{0} in H(\mathbb{T}) if for all compact K\subseteq\mathbb{D} , f_{n}\rightarrow f_{0} uniformly on K . We remark that for all 0<p , H^{p}(\mathbb{T})\subseteq N^{+}\subseteq N\subseteq H(\mathbb{T}) , where N^{+} is the Smirnov class and N is the Nevanlinna class. For a set of analytic functions on the disc X , we will the write \overline{X} to mean the conjugate of all functions in X , i.e. \overline{X}=\{\overline{f(z)}:f(z)\in X\} . For 1<p<\infty , Section 3.3 in [ 6 ] shows that the Riesz projection is a bounded operator P:L^{p}(\mathbb{T})\to H^{p}(\mathbb{T}) , and consequently for 1<p<\infty , we can write L^{p}(\mathbb{T})=\overline{zH^{p}(\mathbb{T})}\oplus H^{p}(\mathbb{T}) . We also note that Section 3.3 in [ 6 ] demonstrates P(L^{1}(\mathbb{T}))\neq H^{1}(\mathbb{T}) . Throughout, we use the notation B_{\epsilon}(y) to mean to open ball radius \epsilon centred at y . 2.2 Toeplitz operators Definition 2.2 . The Toeplitz operator with symbol g\in L^{2}(\mathbb{T}) is the mapping f\mapsto P(gf) (where here the multiplication gf is understood as multiplication of the functions f and g on \mathbb{T} ). Let u\in H(\mathbb{T}),v\in L^{2}(\mathbb{T}) and let 0<s<1 . Remark 3.4.3 in [ 6 ] shows that P:L^{1}(\mathbb{T})\to H^{s}(\mathbb{T}) is continuous and consequently the map \tilde{T_{{v}}}:H^{2}\to H^{s}(\mathbb{T}) such that \tilde{T}_{v}f=P({v}f) is continuous. Define \tilde{T_{u}}:H^{s}(\mathbb{T})\to H(\mathbb{T}) by f\mapsto uf . As H^{s}(\mathbb{T}) convergence implies convergence in H(\mathbb{T})