Distribution of functions in abstract \(H^ 1\) (Q1892733)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Distribution of functions in abstract \(H^ 1\)
scientific article

    Statements

    Distribution of functions in abstract \(H^ 1\) (English)
    0 references
    0 references
    4 July 1996
    0 references
    Let \(\mathcal M\) be a Polish space with a non-atomic probability measure \(\mu\) and let \(A\) be a weak\(^*\)-Dirichlet algebra on \(({\mathcal M}, \mu)\), i.e. a subalgebra of \(L^\infty(\mu)\) such that \(\mu\) is multiplicative on \(A\), \(A\) contains constants and \(A+ \overline A\) is weak\(^*\)-dense in \(L^\infty(\mu)\). The abstract Hardy spaces \(H^p(\mu)\) are defined to be the closure of \(A\) in \(L^p(\mu)\) for \(1\leq p< \infty\) and the weak\(^*\)-closure of \(A\) in \(L^\infty(\mu)\) for \(p= \infty\). The conjugation operator \(\sim: u\mapsto\widetilde u\) is then defined as a mapping from \(\text{Re }A\) into \(\text{Re }A_0\) such that \(u+ i\widetilde u\in A\), where \(A_0= \{f\in A: \int fd\mu= 0\}\). The conjugation operator is then extended to a bounded operator from \(L^p(\mu)\) to \(L^p(\mu)\) if \(1< p< \infty\). But, when \(p= 1\), it can only be extended to a bounded operator from \(L^1(\mu)\) into \(L^p(\mu)\) for \(0< p< 1\). The problem is to characterize such functions \(f\in L^1(\mu)\) that satisfy \(\widetilde f\in L^1(\mu)\). This problem was solved by \textit{B. Davis} [Trans. Am. Math. Soc. 261, 211-233 (1980; Zbl 0438.42010)] in the case of the unit circle and the Euclidean spaces. In the paper under review, the author gives an answer to the problem in the case of weak\(^*\)-Dirichlet algebras. For any real-valued function \(f\) on \(\mathbb{R}\) let \(f_\delta\) be the signed decreasing function which has the same distribution function as \(f\). We set \(M(f; t)= \int^t_{- t} f_\delta(u) du\). The main theorem states that if a function \(f\in L^1(\mu)\) belongs to \(\text{Re } H^1_0(\mu)\) then \(\int^\infty_0 |M(f; t)|t^{- 1}dt< \infty\). This is a partial extension of a result of Davis mentioned above to the case of weak\(^*\)-Dirichlet algebra. In the proof, the author proves a generalized Jensen inequality and then uses an idea of \textit{N. J. Kalton} [`Nonlinear commutators in interpolation theory', Mem. Am. Math. Soc. 385, 85 p. (1988; Zbl 0658.46059)]. The author's result may be applied to, for instance, algebras of abstract analytic functions on groups with ordered dual as well as ergodic Hardy spaces.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Polish space
    0 references
    non-atomic probability measure
    0 references
    weak\(^*\)-Dirichlet algebra
    0 references
    abstract Hardy spaces
    0 references
    conjugation operator
    0 references
    algebras of abstract analytic functions on groups with ordered dual
    0 references
    ergodic Hardy spaces
    0 references