Skip to Content
MathematicsMathematics
  • Article
  • Open Access

18 November 2026

18 Pages

Products of Extended Cesàro and Composition Operators on Radial Bloch-Type Spaces of Complex Banach Spaces

and
1
Department of Mathematics, College of Science, King Saud University, Riyadh 11421, Saudi Arabia
2
Department of Mathematical Sciences, Talladega College, Talladega, AL 35160, USA
*
Author to whom correspondence should be addressed.

Abstract

We study products of extended Cesàro and composition operators P φ ψ = T ψ C φ between radial Bloch-type spaces on the unit ball of a complex Banach space. We characterize boundedness in terms of the symbols φ and ψ and establish two-sided operator-norm estimates. Under a relative compactness assumption on the interior images of φ , we characterize compactness through a boundary vanishing condition. In the finite-dimensional setting, we also derive two-sided essential-norm estimates in terms of the corresponding boundary quantity. Applications to classical Bloch, power-type, logarithmic, and iterated logarithmic weights yield explicit integral expressions and corresponding norm estimates.

1. Introduction

Let X be a complex Banach space, with open unit ball B X , and let H ( B X ) denote the holomorphic functions on B X . For f ∈ H ( B X ) , we write
R f ( z ) = D f ( z ) z ,
where D f ( z ) is the Fréchet derivative. Given a positive continuous weight ω on [ 0 , 1 ) , the radial Bloch-type space B ω R ( B X ) consists of those f ∈ H ( B X ) for which
∥ f ∥ B ω R = | f ( 0 ) | + sup z ∈ B X ω ( ∥ z ∥ ) | R f ( z ) | < ∞ .
We work with normal weights: a positive continuous weight ω is normal if there are 0 < a < b and δ ∈ [ 0 , 1 ) such that ω ( r ) / ( 1 − r ) a decreases to zero and ω ( r ) / ( 1 − r ) b increases to infinity as r → 1 − , on [ δ , 1 ) . Integral operators have long been studied on Hardy, Bergman, and Bloch-type spaces. Early work on Cesàro-type and related operators includes [1,2,3,4]. On finite-dimensional balls, Hu [5], Wang and Hu [6], and Tang [7] obtained results for extended Cesàro operators on Bloch and Bloch-type spaces. Hamada [8] subsequently studied these operators on radial Bloch-type spaces over the unit ball of a complex Banach space, developing estimates and auxiliary functions suited to normal weights. For ψ ∈ H ( B X ) , the extended Cesàro operator is
T ψ f ( z ) = ∫ 0 1 f ( t z ) R ψ ( t z ) d t t .
If φ : B X → B X is holomorphic, let C φ f = f ∘ φ . We consider the product in the specified order
P φ ψ = T ψ C φ ,
so that
P φ ψ f ( z ) = ∫ 0 1 f ( φ ( t z ) ) R ψ ( t z ) d t t .
The radial derivative has the particularly useful form
R ( P φ ψ f ) ( z ) = f ( φ ( z ) ) R ψ ( z ) .
Thus the mapping properties of the product depend both on evaluation at φ ( z ) and on the radial derivative of ψ . Products of composition and integral operators are not new. Li and Stević [9] treated Volterra-type products on classical analytic function spaces; Liang and Zhou [10] and Yang [11] studied related operators between other holomorphic function spaces. More directly relevant here, Quang [12] investigated extended Cesàro composition operators on weak Bloch-type spaces over Hilbert balls, and Lam and Quang [13] obtained boundedness, compactness, and norm estimates for related operators on finite-dimensional balls. These results make it important to distinguish the setting and hypotheses of the present work from those already considered. Our focus is the product P φ ψ acting between radial Bloch-type spaces on a general complex Banach ball. The infinite-dimensional setting matters most in the compactness argument: a norm-bounded subset of B X need not have compact closure, even when it stays a positive distance from the boundary. We therefore assume, for the compactness characterization, that the interior images
E ρ φ = { φ ( z ) : z ∈ B X , ∥ φ ( z ) ∥ ≤ ρ }
are relatively compact for each 0 < ρ < 1 . This condition is automatic in finite dimensions. In a general Banach space it supplies the compact sets needed for a diagonal subsequence argument; we do not claim it is necessary for compactness in every case. We establish necessary and sufficient conditions for boundedness of the product operator between radial Bloch-type spaces, together with two-sided operator-norm estimates. Under the stated weight and relative compactness hypotheses, we characterize compactness through a boundary-vanishing condition. We also obtain two-sided essential-norm estimates in the finite-dimensional setting, where compact dilation operators are available. The estimates use weight-dependent point-evaluation bounds rather than exact evaluation-norm formulas. These conclusions extend the analysis to the radial Banach-space framework while retaining explicit limits on what the compactness and essential-norm arguments establish. We illustrate the criteria for classical Bloch, power-type, logarithmic, and iterated logarithmic weights. Section 2 treats boundedness and operator norms; Section 3 addresses compactness; Section 4 considers essential norms; and Section 5 gives the weight examples.

2. Boundedness of P φ ψ

In this section, we characterize the boundedness of the extended Cesàro composition operator
P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) ,
where ω and μ are normal weights. We also obtain an equivalent estimate for the operator norm. We first record the following point-evaluation estimate.
Lemma 1. 
Let ω be a normal weight. Then, there exists a constant K ω > 0 , depending only on ω, such that for every f ∈ B ω R ( B X ) and z ∈ B X ,
| f ( z ) | ≤ K ω max 1 , ∫ 0 ∥ z ∥ d t ω ( t ) ∥ f ∥ B ω R ,
where K ω = 2 + 1 m ω .
Proof. 
Set m ω = min 0 ≤ t ≤ 1 / 2 ω ( t ) > 0 , and let M = sup w ∈ B X ω ( ∥ w ∥ ) | R f ( w ) | . Fix z ∈ B X ∖ { 0 } . By the radial integration formula,
f ( z ) − f ( 0 ) = ∫ 0 1 R f ( s z ) s d s .
We first consider ∥ z ∥ > 1 / 2 . Splitting the integral gives
| f ( z ) − f ( 0 ) | ≤ ∫ 0 1 / ( 2 ∥ z ∥ ) | R f ( s z ) | s d s + ∫ 1 / ( 2 ∥ z ∥ ) 1 | R f ( s z ) | s d s .
To estimate the first integral, observe that the holomorphic function
λ ⟼ R f λ z 2 ∥ z ∥ , | λ | < 1 ,
vanishes at zero and is bounded in modulus by M / m ω . Schwarz’s lemma therefore yields
| R f ( s z ) | ≤ 2 M m ω s ∥ z ∥ , s ∥ z ∥ ≤ 1 2 .
Consequently,
∫ 0 1 / ( 2 ∥ z ∥ ) | R f ( s z ) | s d s ≤ M m ω .
For the second integral, the definition of M gives
∫ 1 / ( 2 ∥ z ∥ ) 1 | R f ( s z ) | s d s ≤ M ∫ 1 / ( 2 ∥ z ∥ ) 1 d s s ω ( s ∥ z ∥ ) = M ∫ 1 / 2 ∥ z ∥ d t t ω ( t ) ≤ 2 M ∫ 0 ∥ z ∥ d t ω ( t ) .
Combining these estimates, we obtain
| f ( z ) − f ( 0 ) | ≤ M m ω + 2 M ∫ 0 ∥ z ∥ d t ω ( t ) .
If 0 < ∥ z ∥ ≤ 1 / 2 , Schwarz’s estimate applies throughout the radial segment, so
| f ( z ) − f ( 0 ) | ≤ ∫ 0 1 2 M m ω ∥ z ∥ d s ≤ M m ω .
Thus, for every z ∈ B X ,
| f ( z ) | ≤ | f ( 0 ) | + M 1 m ω + 2 ∫ 0 ∥ z ∥ d t ω ( t ) .
Since ∥ f ∥ B ω R = | f ( 0 ) | + M , it follows that
| f ( z ) | ≤ K ω max 1 , ∫ 0 ∥ z ∥ d t ω ( t ) ∥ f ∥ B ω R .
This completes the proof. □
We next record the test-function construction of Hamada [8] (Lemmas 2.1, 2.6 and 2.7). We state the precise properties needed below because they are essential in both the boundedness and compactness arguments.
Lemma 2. 
Let ω be a normal weight. Denote
k 0 = max 0 , log 2 1 ω ( δ ) ,
r k = ω | [ δ , 1 ) − 1 1 2 k , n k = 1 1 − r k , k > k 0 ,
where [ x ] denotes the greatest integer not exceeding x, and the inverse is taken on the strictly decreasing restriction ω | [ δ , 1 ) . For k > k 0 , the number 2 − k lies in the range of this restriction, so r k is well defined. Let
g ( ζ ) = 1 + ∑ k > k 0 ∞ 2 k ζ n k , ζ ∈ D .
Then,
(i) 
g is holomorphic on D , g ( r ) is increasing on [ 0 , 1 ) , and
0 < C 1 = inf r ∈ [ 0 , 1 ) ω ( r ) g ( r ) ≤ sup r ∈ [ 0 , 1 ) ω ( r ) g ( r ) = C 2 < ∞ .
(ii) 
There exists a constant C 3 > 0 such that
∫ 0 r g ( t ) d t ≤ C 3 ∫ 0 r 2 g ( t ) d t , r ∈ [ r 1 , 1 ) ,
where r 1 ∈ ( 0 , 1 ) is the unique number satisfying ∫ 0 r 1 g ( t ) d t = 1 . Such a number exists because g ( t ) > 1 for t > 0 , so ∫ 0 1 g ( t ) d t > 1 . The comparison estimate and C 3 are as in Hamada [8] (Lemma 2.1(ii)).
Lemma 3. 
Let g ∈ H ( D ) be the function defined in Lemma 2. For each v ∈ B X ∖ { 0 } , set
T ( v ) = l v ∈ X * : l v ( v ) = ∥ v ∥ , ∥ l v ∥ = 1 .
For l v ∈ T ( v ) , define
f v ( z ) = ∫ 0 ∥ v ∥ l v ( z ) g ( ζ ) d ζ , z ∈ B X .
Then, f v ∈ B ω , 0 R ( B X ) and ∥ f v ∥ B ω R ≤ C 2 , where C 2 is the constant in Lemma 2.
Theorem 1. 
Let X be a complex Banach space, let ω and μ be normal weights, let φ : B X → B X be holomorphic, and let ψ ∈ H ( B X ) . Then P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded if and only if
M φ , ψ : = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) < ∞ .
Moreover,
1 C ω M φ , ψ ≤ ∥ P φ ψ ∥ ≤ K ω M φ , ψ ,
where
C ω = max 1 , 1 C 1 , C 2 C 3 C 1 .
Proof. 
Suppose first that (7) holds. Let f ∈ B ω R ( B X ) . By (2),
R ( P φ ψ f ) ( z ) = f ( φ ( z ) ) R ψ ( z ) .
By Lemma 1,
| f ( φ ( z ) ) | ≤ K ω max 1 , ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ∥ f ∥ B ω R .
Therefore,
μ ( ∥ z ∥ ) | R ( P φ ψ f ) ( z ) | ≤ K ω μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ∥ f ∥ B ω R ≤ K ω M φ , ψ ∥ f ∥ B ω R .
Since
P φ ψ f ( 0 ) = 0 ,
we obtain
∥ P φ ψ f ∥ B μ R ≤ K ω M φ , ψ ∥ f ∥ B ω R .
Thus, P φ ψ is bounded and
∥ P φ ψ ∥ ≤ K ω M φ , ψ .
Conversely, suppose that P φ ψ is bounded. Taking f ≡ 1 , we have
∥ f ∥ B ω R = 1 ,
and (2) gives
μ ( ∥ z ∥ ) | R ψ ( z ) | = μ ( ∥ z ∥ ) | R ( P φ ψ f ) ( z ) | ≤ ∥ P φ ψ ∥ .
Hence,
μ ( ∥ z ∥ ) | R ψ ( z ) | ≤ ∥ P φ ψ ∥ , z ∈ B X .
We next estimate the integral term. Fix z ∈ B X and set v = φ ( z ) . Let r 1 be the constant appearing in Lemma 2. Suppose first that ∥ v ∥ ≥ r 1 . Let f v be the function defined in Lemma 3. Then,
f v ( v ) = ∫ 0 ∥ v ∥ 2 g ( t ) d t .
Since ω ( t ) g ( t ) ≥ C 1 , we have 1 ω ( t ) ≤ g ( t ) C 1 . Therefore, by Lemma 2,
∫ 0 ∥ v ∥ d t ω ( t ) ≤ 1 C 1 ∫ 0 ∥ v ∥ g ( t ) d t ≤ C 3 C 1 ∫ 0 ∥ v ∥ 2 g ( t ) d t = C 3 C 1 f v ( v ) .
Using (2) and ∥ f v ∥ B ω R ≤ C 2 , we obtain
μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ C 3 C 1 μ ( ∥ z ∥ ) | R ψ ( z ) | | f v ( v ) | = C 3 C 1 μ ( ∥ z ∥ ) | R ( P φ ψ f v ) ( z ) | ≤ C 2 C 3 C 1 ∥ P φ ψ ∥ .
Now suppose that ∥ v ∥ < r 1 . Since 1 ω ( t ) ≤ g ( t ) C 1 and ∫ 0 r 1 g ( t ) d t = 1 , we obtain
∫ 0 ∥ v ∥ d t ω ( t ) ≤ 1 C 1 ∫ 0 r 1 g ( t ) d t = 1 C 1 .
Therefore, by (10),
μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ 1 C 1 ∥ P φ ψ ∥ .
Consequently, for every z ∈ B X ,
μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ C ω ∥ P φ ψ ∥ ,
where
C ω = max 1 , 1 C 1 , C 2 C 3 C 1 .
Taking the supremum over z ∈ B X yields
M φ , ψ ≤ C ω ∥ P φ ψ ∥ .
Thus,
1 C ω M φ , ψ ≤ ∥ P φ ψ ∥ .
Combining (9) and (11), we obtain
1 C ω M φ , ψ ≤ ∥ P φ ψ ∥ ≤ K ω M φ , ψ .
The proof is complete. □
Remark 1. 
Suppose that
∥ φ ∥ ∞ : = sup z ∈ B X ∥ φ ( z ) ∥ < 1 .
Then,
∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ ∫ 0 ∥ φ ∥ ∞ d t ω ( t ) < ∞ .
Consequently, the boundedness condition for P φ ψ reduces to
sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | < ∞ .
Thus, in this case, P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded if and only if ψ ∈ B μ R ( B X ) .
Corollary 1. 
Let ω and μ be normal weights, and let ψ ∈ H ( B X ) . Define
M ψ : = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , ∫ 0 ∥ z ∥ d t ω ( t ) .
Then, T ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded if and only if M ψ < ∞ . Moreover,
1 C ω M ψ ≤ ∥ T ψ ∥ ≤ K ω M ψ ,
where
C ω = max 1 , 1 C 1 , C 2 C 3 C 1 .
Proof. 
The result follows from Theorem 1 by taking φ ( z ) = z . □

3. Compactness of P φ ψ

In this section, we study the compactness of the operator P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) . Since X may be infinite-dimensional, an additional relative compactness condition is needed. For 0 < ρ < 1 , define
E ρ φ = φ ( z ) : z ∈ B X , ∥ φ ( z ) ∥ ≤ ρ .
We assume that E ρ φ is relatively compact in B X for every 0 < ρ < 1 . This hypothesis is used in the sufficiency proof to ensure that E ρ φ ¯ is compact in B X for every 0 < ρ < 1 . In infinite-dimensional spaces, we do not claim that this hypothesis is necessary for the compactness of P φ ψ .
We use the following compactness criterion formulated in terms of bounded nets, following the approach of Blasco et al. [14] (Lemma 4.4). The proof is straightforward and is therefore omitted.
Lemma 4. 
Let X be a complex Banach space, let ω and μ be normal weights, let φ : B X → B X be holomorphic, and let ψ ∈ H ( B X ) . Suppose that P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded. Then P φ ψ is compact if and only if, for every bounded net ( f α ) in B ω R ( B X ) such that f α → 0 uniformly on compact subsets of B X , we have
∥ P φ ψ f α ∥ B μ R ⟶ 0 .
Lemma 5 
([8] (Lemma 2.7)). For v ∈ B X with ∥ v ∥ ≥ r 1 , let
F v ( z ) = ( f v ( z ) ) 2 f v ( v ) , z ∈ B X .
Then, F v ∈ B ω , 0 R ( B X ) and
∥ F v ∥ B ω R ≤ 2 C 2 C 3 .
Moreover, if ∫ 0 1 d t ω ( t ) = ∞ , then F v converges to 0 uniformly on any closed ball strictly inside B X as ∥ v ∥ → 1 .
We now give the compactness characterization.
Theorem 2. 
Let X be a complex Banach space, let ω and μ be normal weights, let φ : B X → B X be holomorphic, and let ψ ∈ H ( B X ) . Suppose that
∫ 0 1 d t ω ( t ) = ∞
and that the relative compactness assumption above holds. Then P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is compact if and only if it is bounded and
lim ∥ φ ( z ) ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) = 0 .
Proof. 
Suppose that P φ ψ is bounded and that
μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ⟶ 0 as ∥ φ ( z ) ∥ → 1 .
Let ( f α ) be a bounded net in B ω R ( B X ) such that f α → 0 uniformly on compact subsets of B X , and put
M : = sup α ∥ f α ∥ B ω R .
Let ε > 0 . Since
∫ 0 1 d t ω ( t ) = ∞
and the boundary condition holds, we can choose ρ ∈ ( 0 , 1 ) sufficiently close to 1 such that
∫ 0 ρ d t ω ( t ) ≥ 1
and
sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ ε .
For ∥ φ ( z ) ∥ > ρ , we have
∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≥ ∫ 0 ρ d t ω ( t ) ≥ 1 .
Hence, by Lemma 1,
sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f α ( φ ( z ) ) | ≤ K ω M sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ≤ K ω M ε .
Therefore,
∥ P φ ψ f α ∥ B μ R ≤ sup ∥ φ ( z ) ∥ ≤ ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f α ( φ ( z ) ) | + sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f α ( φ ( z ) ) | ≤ sup ∥ φ ( z ) ∥ ≤ ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f α ( φ ( z ) ) | + K ω M ε .
Since P φ ψ is bounded,
A : = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | < ∞ .
By the relative compactness assumption, K ρ : = E ρ φ ¯ is compact in B X . Hence,
sup ∥ φ ( z ) ∥ ≤ ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f α ( φ ( z ) ) | ≤ A sup w ∈ K ρ | f α ( w ) | ⟶ 0 .
Consequently,
lim sup α ∥ P φ ψ f α ∥ B μ R ≤ K ω M ε .
Since ε > 0 is arbitrary,
∥ P φ ψ f α ∥ B μ R ⟶ 0 .
Thus, P φ ψ is compact by Lemma 4.
Conversely, suppose that P φ ψ is compact. Then it is bounded. Let ( z j ) ⊂ B X satisfy ∥ φ ( z j ) ∥ ⟶ 1 , and put v j = φ ( z j ) . After discarding finitely many terms, we may assume that ∥ v j ∥ ≥ r 1 . Set f j = F v j . By Lemma 5, f j ∈ B ω , 0 R ( B X ) , ∥ f j ∥ B ω R ≤ 2 C 2 C 3 , and f j ⟶ 0 uniformly on compact subsets of B X . By the compactness of P φ ψ and Lemma 4, ∥ P φ ψ f j ∥ B μ R ⟶ 0 .
On the other hand,
∥ P φ ψ f j ∥ B μ R ≥ μ ( ∥ z j ∥ ) R ( P φ ψ f j ) ( z j ) = μ ( ∥ z j ∥ ) | R ψ ( z j ) | F v j ( v j ) .
Since
F v j ( v j ) = ∫ 0 ∥ v j ∥ 2 g ( t ) d t
and ω ( t ) g ( t ) ≥ C 1 , Lemma 2 gives
∫ 0 ∥ v j ∥ d t ω ( t ) ≤ 1 C 1 ∫ 0 ∥ v j ∥ g ( t ) d t ≤ C 3 C 1 ∫ 0 ∥ v j ∥ 2 g ( t ) d t = C 3 C 1 F v j ( v j ) .
Therefore,
F v j ( v j ) ≥ C 1 C 3 ∫ 0 ∥ v j ∥ d t ω ( t ) .
Thus, by (15),
∥ P φ ψ f j ∥ B μ R ≥ C 1 C 3 μ ( ∥ z j ∥ ) | R ψ ( z j ) | ∫ 0 ∥ v j ∥ d t ω ( t ) = C 1 C 3 μ ( ∥ z j ∥ ) | R ψ ( z j ) | ∫ 0 ∥ φ ( z j ) ∥ d t ω ( t ) .
Since the left-hand side tends to zero, we conclude that
μ ( ∥ z j ∥ ) | R ψ ( z j ) | ∫ 0 ∥ φ ( z j ) ∥ d t ω ( t ) ⟶ 0 .
□
Remark 2. 
Suppose that
∥ φ ∥ ∞ : = sup z ∈ B X ∥ φ ( z ) ∥ < 1 .
Then, the boundary condition in the compactness theorem is automatically satisfied, since ∥ φ ( z ) ∥ cannot approach 1. Consequently, if φ ( B X ) is relatively compact in B X , then P φ ψ is compact whenever it is bounded. In view of the boundedness criterion, in this case we have
P φ ψ is compact ⟺ P φ ψ is bounded ⟺ ψ ∈ B μ R ( B X ) .
Corollary 2. 
Let X be a finite-dimensional complex Banach space, and let ω and μ be normal weights satisfying
∫ 0 1 d t ω ( t ) = ∞ .
Let ψ ∈ H ( B X ) and suppose that T ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded. Then, T ψ is compact if and only if
lim ∥ z ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ z ∥ d t ω ( t ) = 0 .
Proof. 
Take φ = id B X in Theorem 2. Since X is finite dimensional, for every 0 < ρ < 1 the set
{ z ∈ B X : ∥ z ∥ ≤ ρ }
is compact. Hence, the relative compactness hypothesis in Theorem 2 is satisfied. □
Remark 3. 
The finite-dimensional assumption in Corollary 2 ensures that the relative compactness hypothesis of the preceding theorem is satisfied when φ ( z ) = z . Hamada [8] (Theorem 4.2(ii)) obtained a compactness characterization for extended Cesàro operators on general complex Banach spaces under a different relative compactness condition. Specifically, for every ε > 0 and 0 < ρ < 1 , the sets
E ε , ρ ψ = z ∈ B X : ∥ z ∥ ≤ ρ , e x i s t s s ≥ 1 , ∥ s z ∥ < 1 , m u ( ∥ s z ∥ ) | R ψ ( s z ) | ≥ ε
are assumed to be relatively compact in B X . Thus, Hamada’s result does not require X to be finite-dimensional, but imposes an additional condition involving the weighted radial derivative of the symbol.

4. Essential Norm of P φ ψ

In this section, we obtain estimates for the essential norm of P φ ψ . Recall that the essential norm of a bounded linear operator T : B ω R ( B X ) ⟶ B μ R ( B X ) is defined by
∥ T ∥ e = inf ∥ T − K ∥ : K : B ω R ( B X ) → B μ R ( B X ) is compact .
For 0 < ρ < 1 , set
M ρ : = sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ φ ( z ) ∥ d t ω ( t ) ,
where the supremum is understood to be zero if the set { z ∈ B X : ∥ φ ( z ) ∥ > ρ } is empty. Also, set
Λ φ , ψ : = lim ρ → 1 − M ρ .
For 0 < r < 1 , define the dilation operator I r f ( z ) = f ( r z ) , z ∈ B X . We first establish its compactness in finite dimensions by a direct subsequence argument.
Lemma 6. 
Let X be a finite-dimensional complex Banach space and let ω be a normal weight. Then, for each 0 < r < 1 , I r is a compact operator on B ω R ( B X ) .
Proof. 
Fix 0 < r < 1 and let ( f n ) be a bounded sequence in B ω R ( B X ) . Since X is finite-dimensional, every closed bounded subset of X is compact. By the point-evaluation estimate and Cauchy estimates, ( f n ) is locally uniformly bounded and equicontinuous. Montel’s theorem therefore yields a subsequence ( f n k ) converging uniformly on compact subsets of B X to a holomorphic function f. Moreover, Cauchy estimates imply that D f n k → D f uniformly on the compact ball { w : ∥ w ∥ ≤ r } .
Since a normal ω is bounded on [ 0 , 1 ) , we obtain
∥ I r f n k − I r f ∥ B ω R ≤ | f n k ( 0 ) − f ( 0 ) | + r sup 0 ≤ t < 1 ω ( t ) sup ∥ w ∥ ≤ r ∥ D f n k ( w ) − D f ( w ) ∥ ⟶ 0 .
The point-evaluation estimate and Cauchy estimates also imply that I r is bounded. Since every bounded sequence ( f n ) admits a subsequence ( f n k ) for which I r f n k converges in the norm of B ω R ( B X ) , the operator I r is compact. □
To establish the lower essential-norm estimate, we first prove the following weak-convergence lemma, adapting the C 0 -space approach of Miralles and Maletzki [15] to the radial Bloch-type setting.
Lemma 7. 
Let X be a finite-dimensional complex Banach space, and let ω be a positive continuous radial weight on B X . Suppose that ( f j ) is a bounded sequence in B ω , 0 R ( B X ) such that f j ⟶ 0 uniformly on compact subsets of B X . Then,
f j ⇀ 0 in B ω , 0 R ( B X ) .
Consequently, if Y is a Banach space and K : B ω , 0 R ( B X ) ⟶ Y is compact, then
∥ K f j ∥ Y ⟶ 0 .
Proof. 
Set
E 0 = f ∈ B ω , 0 R ( B X ) : f ( 0 ) = 0 .
For f ∈ E 0 , define
J f ( z ) = ω ( ∥ z ∥ ) R f ( z ) , z ∈ B X .
Since f ∈ B ω , 0 R ( B X ) ,
ω ( ∥ z ∥ ) | R f ( z ) | ⟶ 0 as ∥ z ∥ → 1 − .
Moreover, J f is continuous on B X . For every ε > 0 , the boundary-vanishing condition gives 0 < s < 1 such that | J f ( z ) | < ε whenever ∥ z ∥ > s . Hence, { z ∈ B X : | J f ( z ) | ≥ ε } is closed and contained in the compact set { z ∈ X : ∥ z ∥ ≤ s } . Thus, J f ∈ C 0 ( B X ) . Since f ( 0 ) = 0 ,
∥ J f ∥ ∞ = sup z ∈ B X ω ( ∥ z ∥ ) | R f ( z ) | = ∥ f ∥ B ω R .
Thus, J : E 0 → C 0 ( B X ) is a linear isometry and, in particular, is injective. Now define g j = f j − f j ( 0 ) . Then, g j ∈ E 0 . Since f j → 0 uniformly on compact subsets of B X , in particular f j ( 0 ) ⟶ 0 . Hence, g j ⟶ 0 uniformly on compact subsets of B X . Since X is finite dimensional, the Cauchy estimates applied on slightly larger compact balls imply that D f j → 0 uniformly on every compact subset of B X . Therefore, for each z ∈ B X , R g j ( z ) = R f j ( z ) = D f j ( z ) ( z ) ⟶ 0 . Consequently,
J g j ( z ) = ω ( ∥ z ∥ ) R f j ( z ) ⟶ 0 , z ∈ B X .
Since ( f j ) is bounded in B ω , 0 R ( B X ) , there exists a constant M > 0 such that
∥ f j ∥ B ω R ≤ M
for all j. In particular,
| J g j ( z ) | = ω ( ∥ z ∥ ) | R f j ( z ) | ≤ M
for every j and every z ∈ B X . We now prove that
g j ⇀ 0 in E 0 .
Let ℓ ∈ E 0 * . Since J is an isometric embedding, the functional Λ defined on the linear subspace J ( E 0 ) of C 0 ( B X ) by
Λ ( J f ) = ℓ ( f ) , f ∈ E 0 ,
is well defined and bounded. Indeed,
| Λ ( J f ) | = | ℓ ( f ) | ≤ ∥ ℓ ∥ ∥ f ∥ B ω R = ∥ ℓ ∥ ∥ J f ∥ ∞ .
By the Hahn–Banach theorem, Λ extends, with the same norm, to a bounded linear functional
Λ ˜ ∈ C 0 ( B X ) * .
By the Riesz–Markov representation theorem, there exists a finite complex regular Borel measure ν on B X , with | ν | ( B X ) < ∞ , such that
Λ ˜ ( h ) = ∫ B X h ( z ) d ν ( z ) , h ∈ C 0 ( B X ) .
Therefore,
ℓ ( g j ) = Λ ( J g j ) = Λ ˜ ( J g j ) = ∫ B X ω ( ∥ z ∥ ) R f j ( z ) d ν ( z ) .
For every z ∈ B X , the integrand converges to zero and its absolute value is bounded by M. Since | ν | ( B X ) < ∞ , the dominated convergence theorem (with respect to | ν | ) yields
ℓ ( g j ) ⟶ 0 .
Since ℓ ∈ E 0 * was arbitrary, g j ⇀ 0 in E 0 . We now show that f j converges weakly to zero. Let L ∈ B ω , 0 R ( B X ) * . The restriction L | E 0 belongs to E 0 * , and therefore L ( g j ) ⟶ 0 . Since f j = g j + f j ( 0 ) 1 and the constant function 1 belongs to B ω , 0 R ( B X ) , we have
L ( f j ) = L ( g j ) + f j ( 0 ) L ( 1 ) .
Both terms on the right-hand side tend to zero. Thus, L ( f j ) ⟶ 0 . Since L was arbitrary, it follows that
f j ⇀ 0 in B ω , 0 R ( B X ) .
Finally, suppose that K : B ω , 0 R ( B X ) ⟶ Y is compact. Since bounded linear operators preserve weak convergence,
K f j ⇀ 0 in Y .
If ∥ K f j ∥ Y did not tend to zero, there would exist ε > 0 and a subsequence ( f j k ) such that
∥ K f j k ∥ Y ≥ ε
for every k. By compactness of K, a further subsequence ( K f j k m ) would converge in norm to some y ∈ Y . Norm convergence implies weak convergence, whereas
K f j k m ⇀ 0 .
Hence, y = 0 , contradicting
∥ K f j k m ∥ Y ≥ ε .
Therefore,
∥ K f j ∥ Y ⟶ 0 .
□
We can now obtain the two-sided estimate.
Theorem 3. 
Let X be a finite-dimensional complex Banach space, let ω and μ be normal weights, let φ : B X → B X be holomorphic, and let ψ ∈ H ( B X ) . Assume that ∫ 0 1 d t ω ( t ) = ∞ and that P φ ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded. Then,
C 1 2 C 2 C 3 2 Λ φ , ψ ≤ ∥ P φ ψ ∥ e ≤ Λ φ , ψ .
Consequently,
∥ P φ ψ ∥ e ≍ Λ φ , ψ .
Proof. 
We first establish the upper estimate. By Lemma 6, the operator P φ ψ I r is compact for every 0 < r < 1 . Hence,
∥ P φ ψ ∥ e ≤ ∥ P φ ψ − P φ ψ I r ∥ = ∥ P φ ψ ( I − I r ) ∥ .
Fix 0 < ρ < 1 and set A : = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | . By Theorem 1, A < ∞ . Let ∥ f ∥ B ω R ≤ 1 . By (2),
∥ P φ ψ ( I − I r ) f ∥ B μ R = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | | f ( φ ( z ) ) − f ( r φ ( z ) ) | .
If ∥ φ ( z ) ∥ ≤ ρ , then
| f ( φ ( z ) ) − f ( r φ ( z ) ) | = ∫ r 1 R f ( t φ ( z ) ) d t t ≤ ∫ r 1 d t t ω ( t ∥ φ ( z ) ∥ ) .
Let m ρ = min 0 ≤ s ≤ ρ ω ( s ) > 0 . Since t ∥ φ ( z ) ∥ ≤ ρ ,
| f ( φ ( z ) ) − f ( r φ ( z ) ) | ≤ log ( 1 / r ) m ρ .
Consequently,
sup ∥ φ ( z ) ∥ ≤ ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f ( φ ( z ) ) − f ( r φ ( z ) ) | ≤ A log ( 1 / r ) m ρ .
Now suppose that ∥ φ ( z ) ∥ > ρ and put w = φ ( z ) . Then,
| f ( w ) − f ( r w ) | ≤ ∫ r 1 | R f ( t w ) | d t t ≤ ∫ r 1 d t t ω ( t ∥ w ∥ ) = ∫ r ∥ w ∥ ∥ w ∥ d s s ω ( s ) .
Since s ≥ r ∥ w ∥ > r ρ ,
| f ( w ) − f ( r w ) | ≤ 1 r ρ ∫ 0 ∥ w ∥ d s ω ( s ) .
Therefore,
sup ∥ φ ( z ) ∥ > ρ μ ( ∥ z ∥ ) | R ψ ( z ) | | f ( φ ( z ) ) − f ( r φ ( z ) ) | ≤ M ρ r ρ .
Combining (16) and (17), we obtain
∥ P φ ψ ( I − I r ) ∥ ≤ A log ( 1 / r ) m ρ + M ρ r ρ .
Hence,
∥ P φ ψ ∥ e ≤ A log ( 1 / r ) m ρ + M ρ r ρ .
Letting r → 1 − gives
∥ P φ ψ ∥ e ≤ M ρ ρ .
Finally, letting ρ → 1 − yields
∥ P φ ψ ∥ e ≤ Λ φ , ψ .
We next establish the lower estimate. If Λ φ , ψ = 0 , the lower estimate is immediate. Suppose Λ φ , ψ > 0 . By the definition of Λ φ , ψ , choose ( z j ) ⊂ B X such that ∥ φ ( z j ) ∥ ⟶ 1 and
μ ( ∥ z j ∥ ) | R ψ ( z j ) | ∫ 0 ∥ φ ( z j ) ∥ d t ω ( t ) ⟶ Λ φ , ψ .
Set v j = φ ( z j ) , and f j = F v j . Since ∥ v j ∥ → 1 , after discarding finitely many terms we may assume that ∥ v j ∥ ≥ r 1 . By Lemma 5, f j ∈ B ω , 0 R ( B X ) , ∥ f j ∥ B ω R ≤ 2 C 2 C 3 , and f j ⟶ 0 uniformly on compact subsets of B X . Let K : B ω R ( B X ) ⟶ B μ R ( B X ) be any compact operator. Its restriction K 0 = K | B ω , 0 R ( B X ) is compact. By Lemma 7,
∥ K f j ∥ B μ R = ∥ K 0 f j ∥ B μ R ⟶ 0 .
Since ω ( t ) g ( t ) ≥ C 1 and ∥ v j ∥ ≥ r 1 , Lemma 2 gives
∫ 0 ∥ v j ∥ d t ω ( t ) ≤ 1 C 1 ∫ 0 ∥ v j ∥ g ( t ) d t ≤ C 3 C 1 ∫ 0 ∥ v j ∥ 2 g ( t ) d t = C 3 C 1 F v j ( v j ) .
Therefore,
F v j ( v j ) ≥ C 1 C 3 ∫ 0 ∥ v j ∥ d t ω ( t ) .
Using ∥ f j ∥ B ω R ≤ 2 C 2 C 3 and inequality (18), we obtain
2 C 2 C 3 ∥ P φ ψ − K ∥ ≥ ∥ ( P φ ψ − K ) f j ∥ B μ R ≥ ∥ P φ ψ f j ∥ B μ R − ∥ K f j ∥ B μ R ≥ μ ( ∥ z j ∥ ) | R ψ ( z j ) | | f j ( φ ( z j ) ) | − ∥ K f j ∥ B μ R = μ ( ∥ z j ∥ ) | R ψ ( z j ) | F v j ( v j ) − ∥ K f j ∥ B μ R ≥ C 1 C 3 μ ( ∥ z j ∥ ) | R ψ ( z j ) | ∫ 0 ∥ v j ∥ d t ω ( t ) − ∥ K f j ∥ B μ R .
Letting j → ∞ gives
2 C 2 C 3 ∥ P φ ψ − K ∥ ≥ C 1 C 3 Λ φ , ψ .
Therefore,
∥ P φ ψ − K ∥ ≥ C 1 2 C 2 C 3 2 Λ φ , ψ .
Taking the infimum over all compact operators K yields
∥ P φ ψ ∥ e ≥ C 1 2 C 2 C 3 2 Λ φ , ψ .
This completes the proof. □
Remark 4. 
The finite-dimensionality assumption is used in the weak-convergence argument for the test functions in the little radial Bloch-type space. In particular, it makes B X locally compact, permitting the Riesz–Markov representation theorem to be applied to C 0 ( B X ) .
Corollary 3. 
Let X be a finite-dimensional complex Banach space, let ω and μ be normal weights satisfying ∫ 0 1 d t ω ( t ) = ∞ , and let ψ ∈ H ( B X ) . Suppose that T ψ : B ω R ( B X ) ⟶ B μ R ( B X ) is bounded. Then
∥ T ψ ∥ e ≍ lim sup ∥ z ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | ∫ 0 ∥ z ∥ d t ω ( t ) .
Proof. 
Take φ ( z ) = z in Theorem 3. □

5. Examples of Normal Weights

In this section, we apply the preceding results to several important classes of normal weights: the classical Bloch weight, power-type weights, logarithmic weights, and iterated logarithmic weights. We also obtain the corresponding essential-norm estimates under the hypotheses of Theorem 3. For the weights considered below, the point-evaluation bounds give explicit two-sided operator-norm estimates. Every essential-norm assertion in this section assumes that X is finite dimensional and P φ ψ is bounded, as required by Theorem 3. Compactness of the dilation operators I r then follows from Lemma 6; it is not an additional assumption. For w ∈ B X , let
δ w ( f ) = f ( w ) , f ∈ B ω R ( B X ) .
Since P φ ψ f ( 0 ) = 0 and
R ( P φ ψ f ) ( z ) = f ( φ ( z ) ) R ψ ( z ) ,
we have
∥ P φ ψ ∥ = sup ∥ f ∥ B ω R ≤ 1 sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | | f ( φ ( z ) ) | = sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | ∥ δ φ ( z ) ∥ .
Thus, two-sided estimates for ∥ δ w ∥ yield corresponding two-sided estimates for ∥ P φ ψ ∥ .

5.1. The Classical Bloch Weight

Let
ω ( r ) = 1 − r 2 , 0 ≤ r < 1 .
This weight is normal: near the boundary it is comparable to ( 1 − r ) , so the normal-weight exponents can be chosen with 0 < a < 1 < b . The corresponding radial Bloch norm is
∥ f ∥ B ω R = | f ( 0 ) | + sup z ∈ B X ( 1 − ∥ z ∥ 2 ) | R f ( z ) | .
Moreover,
∫ 0 r d t 1 − t 2 = 1 2 log 1 + r 1 − r .
We obtain explicit two-sided estimates for the point-evaluation norm. For w ∈ B X ∖ { 0 } , choose l w ∈ X * such that
∥ l w ∥ = 1 , l w ( w ) = ∥ w ∥ ,
and define
F w ( z ) = ∫ 0 l w ( z ) d ζ 1 − ζ 2 .
Then, F w ( 0 ) = 0 and
R F w ( z ) = l w ( z ) 1 − l w ( z ) 2 .
Since
| 1 − l w ( z ) 2 | ≥ 1 − | l w ( z ) | 2 ≥ 1 − ∥ z ∥ 2 ,
we obtain
( 1 − ∥ z ∥ 2 ) | R F w ( z ) | ≤ | l w ( z ) | ≤ 1 .
Hence, ∥ F w ∥ B ω R ≤ 1 , while
F w ( w ) = 1 2 log 1 + ∥ w ∥ 1 − ∥ w ∥ .
Here, m ω = 3 / 4 and K ω = 10 / 3 . Combining this test function and the constant function f ≡ 1 with Lemma 1, we obtain
max 1 , 1 2 log 1 + ∥ w ∥ 1 − ∥ w ∥ ≤ ∥ δ w ∥ ≤ K ω max 1 , 1 2 log 1 + ∥ w ∥ 1 − ∥ w ∥ .
Therefore,
∥ P φ ψ ∥ ≍ sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , 1 2 log 1 + ∥ φ ( z ) ∥ 1 − ∥ φ ( z ) ∥ .
Since
∫ 0 1 d t 1 − t 2 = ∞ ,
under the hypotheses of Theorem 3,
∥ P φ ψ ∥ e ≍ lim sup ∥ φ ( z ) ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | 1 2 log 1 + ∥ φ ( z ) ∥ 1 − ∥ φ ( z ) ∥ .

5.2. Power-Type Bloch Spaces

Let
ω ( r ) = ( 1 − r ) α , α > 1 .
This is a normal weight: choose 0 < a < α < b . The restriction α > 1 is used here to ensure divergence of the associated growth integral and hence applicability of the essential-norm theorem. The associated radial Bloch-type norm is
∥ f ∥ B ω R = | f ( 0 ) | + sup z ∈ B X ( 1 − ∥ z ∥ ) α | R f ( z ) | .
We have
∫ 0 r d t ( 1 − t ) α = ( 1 − r ) 1 − α − 1 α − 1 .
The point-evaluation norm again admits explicit two-sided estimates. For w ∈ B X ∖ { 0 } , choose l w ∈ X * such that ∥ l w ∥ = 1 , l w ( w ) = ∥ w ∥ , and define
F w ( z ) = ∫ 0 l w ( z ) d ζ ( 1 − ζ ) α .
Then, F w ( 0 ) = 0 and
R F w ( z ) = l w ( z ) ( 1 − l w ( z ) ) α .
Since
| 1 − l w ( z ) | ≥ 1 − | l w ( z ) | ≥ 1 − ∥ z ∥ ,
it follows that
( 1 − ∥ z ∥ ) α | R F w ( z ) | ≤ | l w ( z ) | ≤ 1 .
Thus, ∥ F w ∥ B ω R ≤ 1 , and
F w ( w ) = ( 1 − ∥ w ∥ ) 1 − α − 1 α − 1 .
Here, m ω = 2 − α and K ω = 2 + 2 α . Combining this test function and the constant function f ≡ 1 with Lemma 1, we obtain
max 1 , ( 1 − ∥ w ∥ ) 1 − α − 1 α − 1 ≤ ∥ δ w ∥ ≤ K ω max 1 , ( 1 − ∥ w ∥ ) 1 − α − 1 α − 1 .
Therefore,
∥ P φ ψ ∥ ≍ sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , ( 1 − ∥ φ ( z ) ∥ ) 1 − α − 1 α − 1 .
Since α > 1 ,
∫ 0 1 d t ( 1 − t ) α = ∞ .
Hence, under the hypotheses of Theorem 3,
∥ P φ ψ ∥ e ≍ lim sup ∥ φ ( z ) ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | ( 1 − ∥ φ ( z ) ∥ ) 1 − α .

5.3. Logarithmic Bloch-Type Spaces

Consider
ω ( r ) = ( 1 − r ) log e 1 − r , 0 ≤ r < 1 .
This is a normal weight: for example, any 0 < a < 1 < b gives the required eventual monotonicity, since logarithmic growth is slower than every positive power of ( 1 − r ) − 1 . The corresponding radial Bloch-type norm is
∥ f ∥ B ω R = | f ( 0 ) | + sup z ∈ B X ( 1 − ∥ z ∥ ) log e 1 − ∥ z ∥ | R f ( z ) | .
A direct calculation gives
∫ 0 r d t ( 1 − t ) log e 1 − t = log log e 1 − r .
In this case, the preceding general estimates give
∥ δ w ∥ ≍ max 1 , log log e 1 − ∥ w ∥ .
Consequently,
∥ P φ ψ ∥ ≍ sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , log log e 1 − ∥ φ ( z ) ∥ .
Since
∫ 0 1 d t ( 1 − t ) log e 1 − t = ∞ ,
under the hypotheses of Theorem 3,
∥ P φ ψ ∥ e ≍ lim sup ∥ φ ( z ) ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | log log e 1 − ∥ φ ( z ) ∥ .

5.4. Iterated Logarithmic Bloch Spaces

For k ∈ N , let
ω k ( r ) = ( 1 − r ) ∏ j = 1 k L j ( r ) ,
where
L 1 ( r ) = log a 1 − r , L j ( r ) = log L j − 1 ( r ) , 2 ≤ j ≤ k ,
and a > 1 is chosen sufficiently large so that L j ( r ) > 0 for 1 ≤ j ≤ k + 1 and 0 ≤ r < 1 . For each fixed k, such an a exists. The resulting weight is normal because its slowly varying logarithmic factor lies between any two sufficiently close powers of ( 1 − r ) − 1 near the boundary. The corresponding radial Bloch-type norm is
∥ f ∥ B ω k R = | f ( 0 ) | + sup z ∈ B X ( 1 − ∥ z ∥ ) ∏ j = 1 k L j ( ∥ z ∥ ) | R f ( z ) | .
Since
d d r L k + 1 ( r ) = 1 ( 1 − r ) L 1 ( r ) ⋯ L k ( r ) ,
we have
∫ 0 r d t ω k ( t ) = L k + 1 ( r ) − L k + 1 ( 0 ) .
Equivalently,
∫ 0 r d t ω k ( t ) = log [ k + 1 ] a 1 − r − log [ k + 1 ] a .
Thus,
∥ δ w ∥ ≍ max 1 , log [ k + 1 ] a 1 − ∥ w ∥ − log [ k + 1 ] a .
It follows that
∥ P φ ψ ∥ ≍ sup z ∈ B X μ ( ∥ z ∥ ) | R ψ ( z ) | max 1 , log [ k + 1 ] a 1 − ∥ φ ( z ) ∥ − log [ k + 1 ] a .
Finally,
∫ 0 1 d t ω k ( t ) = ∞ ,
and therefore, under the hypotheses of Theorem 3,
∥ P φ ψ ∥ e ≍ lim sup ∥ φ ( z ) ∥ → 1 μ ( ∥ z ∥ ) | R ψ ( z ) | log [ k + 1 ] a 1 − ∥ φ ( z ) ∥ .

6. Conclusions

We established boundedness and compactness criteria for P φ ψ = T ψ C φ on radial Bloch-type spaces over complex Banach balls, together with operator-norm and essential-norm estimates. Applications to standard normal weights yield explicit integral expressions and two-sided norm estimates. The compactness characterization assumes relative compactness of the interior images of φ , a sufficient condition whose necessity is not established. The two-sided essential-norm estimate requires finite dimensionality. Extending these estimates to infinite-dimensional Banach spaces remains a direction for future research.

Author Contributions

Conceptualization, R.A. and N.H.; writing—original draft, R.A. and N.H.; writing—review and editing, R.A. and N.H. All authors have read and agreed to the published version of the manuscript.

Funding

The authors were supported by Researchers Supporting Project number (RSPD2026R640), King Saud University, Riyadh, Saudi Arabia.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Acknowledgments

The authors extend their appreciation to King Saud University, Saudi Arabia for funding this research work through grant number: RSPD2026R640.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Aleman, A.; Cima, J.A. An integral operator on Hp and Hardy’s inequality. J. Anal. Math. 2001, 85, 157–176. [Google Scholar] [CrossRef] [Scilit]
  2. Aleman, A.; Siskakis, A.G. Integration operators on Bergman spaces. Ind. Univ. Math. J. 1997, 46, 337–356. [Google Scholar] [CrossRef] [Scilit]
  3. Xiao, J. Cesàro-type operators on Hardy, BMOA and Bloch spaces. Arch. Math. 1997, 68, 398–406. [Google Scholar] [CrossRef] [Scilit]
  4. Xiao, J. Riemann–Stieltjes operators on weighted Bloch and Bergman spaces of the unit ball. J. Lond. Math. Soc. 2004, 70, 199–214. [Google Scholar] [CrossRef] [Scilit]
  5. Hu, Z. Extended Cesàro operators on the Bloch space in the unit ball of C n . Acta Math. Sci. Ser. B Engl. Ed. 2003, 23, 561–566. [Google Scholar] [CrossRef] [Scilit]
  6. Wang, S.S.; Hu, Z.J. Generalized Cesàro operators on Bloch-type spaces. Chin. Ann. Math. Ser. A 2005, 26, 613–624. [Google Scholar]
  7. Tang, X. Extended Cesàro operators between Bloch-type spaces in the unit ball of C n . J. Math. Anal. Appl. 2007, 326, 1199–1211. [Google Scholar] [CrossRef] [Scilit]
  8. Hamada, H. Bloch-type spaces and extended Cesàro operators in the unit ball of a complex Banach space. Sci. China Math. 2019, 62, 617–628. [Google Scholar] [CrossRef] [Scilit]
  9. Li, S.; Stević, S. Products of Volterra type operator and composition operator from H∞ and Bloch spaces to Zygmund spaces. J. Math. Anal. Appl. 2008, 345, 40–52. [Google Scholar] [CrossRef] [Scilit]
  10. Liang, Y.-X.; Zhou, Z.-H. Product of extended Cesàro operator and composition operator from Lipschitz space to F(p, q, s) space on the unit ball. Abstr. Appl. Anal. 2011, 2011, 152635. [Google Scholar] [CrossRef] [Scilit]
  11. Yang, W. Volterra composition operators from F(p, q, s) spaces to Bloch-type spaces. Bull. Malays. Math. Sci. Soc. 2011, 34, 267–277. [Google Scholar]
  12. Quang, T.T. Extended Cesàro composition operators on weak Bloch-type spaces on the unit ball of a Hilbert space. arXiv 2022, arXiv:2203.05129. [Google Scholar]
  13. Lam, L.V.; Quang, T.T. On the boundedness and compactness of extended Cesàro composition operators between weighted Bloch-type spaces. Hacet. J. Math. Stat. 2024, 53, 894–914. [Google Scholar] [CrossRef] [Scilit]
  14. Blasco, O.; Galindo, P.; Lindström, M.; Miralles, A. Composition operators on the Bloch space of the unit ball of a Hilbert space. Banach J. Math. Anal. 2017, 11, 311–334. [Google Scholar] [CrossRef] [Scilit]
  15. Miralles, A.; Maletzki, M.P. On interpolating sequences for Bloch type spaces. J. Math. Anal. Appl. 2022, 511, 126079. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.