Bohr phenomenon for certain integral operators and transforms in complex Banach spaces Vasudevarao Allu, Raju Biswas, Rajib Mandal and Hiroshi Yanagihara Abstract . In this paper, we investigate several Bohr radii associated with the Cesáro operator, Bernardi integral operator, β \beta -Cesáro operator, and discrete Fourier transform, all defined on a set of holomorphic mappings from the unit ball of a complex Banach space into the closure of the unit polydisc 𝔻 n \mathbb{D}^{n} within the space ℂ n \mathbb{C}^{n} . † † footnotetext: 2020 Mathematics Subject Classification: 32A05, 32A10, 32K05, 32M15. † † footnotetext: Key words and phrases: Bohr radius, holomorphic function, Homogeneous polynomial expansion, Cesáro operator, Bernardi integral operator, discrete Fourier transform, \beta -Cesáro operator. 1. Introduction and Preliminaries Let f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} be a holomorphic mapping in the unit disk \mathbb{D}:=\{z\in\mathbb{C}:|z|<1\} . The classical Bohr’s theorem states that if |f(z)|\leq 1 in \mathbb{D} , then \sum_{n=0}^{\infty}|a_{n}|r^{n}\leq 1\quad\text{for all}\quad|z|=r\leq 1/3. (1.1) The number 1/3 cannot be improved and is known as the Bohr radius (see [ 13 ] ). The inequality ( 1.1 ) is known as the classical Bohr inequality. In fact, Bohr’s [ 13 ] original proof of the inequality ( 1.1 ) was limited to the case when r was no greater than 1/6 . Subsequently, Wiener, Schur, and Riesz (see [ 42 , 43 ] ) independently proved that the inequality ( 1.1 ) holds for r\leq 1/3 . Bohr’s and Wiener’s proofs can be found in [ 13 ] . In the recent years, there has been a significant increase in research studies concerning Bohr-type theorems in various abstract settings, including refinements, ramifications, and extensions. For a detailed study on the Bohr radius in one complex variable, we refer to [ 6 , 26 , 3 , 2 , 7 , 8 , 4 , 14 , 23 , 33 , 35 , 5 , 12 , 38 , 28 , 40 , 9 , 10 , 22 ] and the references therein. A natural question arises: “Is it possible to prove the Bohr-type inequality for certain complex integral operators and integral transforms defined on different function spaces?” The idea was initially proposed in the context of the unit disk \mathbb{D} for the classical Cesáro operator in [ 25 , 26 , 27 ] and for the Bernardi integral operator in [ 30 ] . The Cesáro operator is studied in [ 21 ] and is defined as follows: \displaystyle\mathcal{C}f(z):=\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}a_{k}\right)z^{n}=\int_{0}^{1}\frac{f(tz)}{1-tz}\;dt (1.2) for a holomorphic function f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} in the unit disk \mathbb{D} . Evidently, \displaystyle|\mathcal{C}f(z)|=\left|\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}a_{k}\right)z^{n}\right|\leq\frac{1}{r}\log\frac{1}{1-r}\quad\text{for each}\quad|z|=r<1. For the Cesáro operator, Kayumov et al. [ 25 ] have established the following Bohr-type inequality. Theorem A . [ 25 , Theorem 1, p. 616] Let f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} be holomorphic in \mathbb{D} and |f(z)|\leq 1 in \mathbb{D} , then \displaystyle\mathcal{C}_{f}(r)=\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}|a_{k}|\right)r^{n}\leq\frac{1}{r}\log\frac{1}{1-r}\quad\text{for}\quad|z|=r\leq R_{1}, where R_{1}(\approx 0.5335) is the positive root of the equation 2x=3(1-x)\log(1/(1-x)) . The number R_{1} cannot be improved. For a holomorphic function f(z)=\sum_{n=m}^{\infty}a_{n}z^{n} in the unit disk \mathbb{D} , the Bernardi integral operator (see [ 36 , p. 11] ) is defined as follows: \displaystyle\mathcal{L}_{\beta}f(z):=\frac{1+\beta}{z^{\beta}}\int_{0}^{z}f(\eta)\eta^{\beta-1}d\eta=(1+\beta)\sum_{n=m}^{\infty}\frac{a_{n}}{\beta+n}z^{n}, (1.3) where \beta>-m and m\geq 0 is an integer. In 2021, Kumar and Sahoo [ 30 ] studied the following Bohr-type inequality for the Bernardi integral operator. Theorem B . [ 30 ] Let \beta>-m . If f(z)=\sum_{n=m}^{\infty}a_{n}z^{n} is holomorphic in \mathbb{D} and |f(z)|\leq 1 in \mathbb{D} , then \displaystyle\sum_{n=m}^{\infty}\frac{|a_{n}|}{\beta+n}r^{n}\leq\frac{r^{m}}{m+\beta}\quad\text{for}\quad|z|=r\leq R_{2}(\beta), where R_{2}(\beta) is the positive root of the equation \displaystyle\frac{x^{m}}{m+\beta}-2\sum_{n=m+1}^{\infty}\frac{x^{n}}{n+\beta}=0. The number R_{2}(\beta) cannot be improved. Let \left\{x_{n}\right\}_{n=0}^{N-1} be a sequence of complex numbers. The discrete Fourier transform (see [ 11 , Chapter 6] ) is defined by \displaystyle\mathcal{F}(x_{n})=\sum_{n=0}^{N-1}x_{n}e^{-2\pi ink/N}. For a holomorphic function f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} in \mathbb{D} with |f(z)|\leq 1 in \mathbb{D} , we perform the discrete Fourier transform on the coefficients a_{k} from k=0 to n which gives \displaystyle\mathcal{F}[f](z)=\sum_{n=0}^{\infty}\left(\sum_{k=0}^{n}a_{k}e^{-2\pi ink/(n+1)}\right)z^{n}. To obtain the Bohr-type inequality, we denote the majorant series of \mathcal{F}[f](z) as \displaystyle\mathcal{F}_{f}(r):=\sum_{n=0}^{\infty}\left(\sum_{k=0}^{n}|a_{k}|\right)r^{n},\quad\text{where}\quad|z|=r<1. (1.4) For the discrete Fourier transform, Ong and Ng [ 39 ] have established the following Bohr-type inequality. Theorem C . [ 39 ] If f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} is holomorphic in \mathbb{D} and |f(z)|\leq 1 in \mathbb{D} , then \displaystyle\mathcal{F}_{f}(r)\leq\frac{1}{1-r}\quad\text{for}\quad r\leq\frac{1}{3}. The constant 1/3 cannot be improved. Motivated by Theorem A, Kumar and Sahoo [ 30 ] have considered the \beta -Cesáro operator ( \beta>0 ), which is defined as \mathcal{C}_{\beta}^{f}(z):=\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}a_{k}\right)z^{n}=\int_{0}^{1}\frac{f(tz)}{(1-tz)^{\beta}}dt=\frac{1}{z}\int_{0}^{z}\frac{f(t)}{(1-t)^{\beta}}dt (1.5) for a holomorphic function f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} in the unit disk \mathbb{D} . Indeed, Kumar and Sahoo [ 30 ] have established the following Bohr-type inequality for the \beta -Cesáro operator. Theorem D . [ 30 ] Let 0<\beta(\not=1) . If f(z)=\sum_{n=0}^{\infty}a_{n}z^{n} is holomorphic in \mathbb{D} and |f(z)|\leq 1 in \mathbb{D} , then \displaystyle\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}|a_{k}|\right)r^{n}\leq\frac{1}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right) for |z|=r\leq R_{3}(\beta) , where R_{3}(\beta) is the positive root of the equation \displaystyle\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}=0. The number R_{3}(\beta) cannot be improved. In particular, if we put f(t)\equiv 1 , f(t)\equiv\sum_{n=1}^{\infty}t^{n} and f(t)\equiv\sum_{n=0}^{\infty}t^{n} in ( 1.5 ), then we obtain, respectively, the following relations: \displaystyle\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}\right)z^{n}=\frac{1}{z}\int_{0}^{z}\frac{1}{(1-t)^{\beta}}dt, (1.6) \displaystyle\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\right)z^{n}=\frac{1}{z}\int_{0}^{z}\frac{t}{(1-t)^{\beta+1}}dt (1.7) and \displaystyle\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\right)z^{n}=\frac{1}{z}\int_{0}^{z}\frac{1}{(1-t)^{\beta+1}}dt. (1.8) From ( 1.6 ) and ( 1.8 ), we obtain the following identity \displaystyle\frac{\Gamma(n+\beta+1)}{\Gamma(n+1)\Gamma(\beta+1)}\equiv\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}. (1.9) Before proceeding with the discussion, it is essential to introduce the requisite notations. For z=(z_{1},z_{2},\cdots,z_{n})\in\mathbb{C}^{n} , let \|z\|_{\infty}=\max\limits_{j=1,2,\cdots,n}|z_{j}| denote the maximum norm. A ploydisc in \mathbb{C}^{n} is the product of discs. The open polydisc with center \alpha=(\alpha_{1},\alpha_{2},\ldots,\alpha_{n}) and poly-radius r=(r_{1},r_{2},\ldots,r_{n}) will be denoted by \mathbb{D}^{n}(\alpha;r) and defined as \displaystyle\mathbb{D}^{n}(\alpha;r)=\left\{z=(z_{1},z_{2},\cdots,z_{n})\in\mathbb{C}^{n}:|z_{j}-\alpha_{j}|R , the following inequality holds \displaystyle\frac{3(1-r)\log(1-r)+2r}{r(1-r)}>0. (2.5) From ( 2.3 ), ( 2.4 ) and ( 2.5 ), we conclude that R cannot be improved. This completes the proof. ∎ Before proceeding to our next result, it is necessary to establish the majorant series corresponding to the Bernardi operator for a holomorphic mapping F:B_{X}\to\overline{\mathbb{D}^{n}} with F(z)=\sum_{s=m}^{\infty}P_{s}(z) , where m\in\mathbb{N}\cup\{0\} , z\in B_{X} , P_{s}(z)=(1/s!)D^{s}F(0)(z^{s}) and F(z)=(F_{1}(z),F_{2}(z),\ldots,F_{n}(z)) . We consider the mapping F_{j}:B_{X}\to\overline{\mathbb{D}} defined by \displaystyle F_{j}(z)=\sum_{s=m}^{\infty}(P_{s})_{j}(z),\quad\text{where}\quad z\in B_{X}\quad\text{and}\quad j\in\{1,2\ldots,n\}. In view of the relation ( 1.3 ), we consider the following integral for the holomorphic mapping F_{j}:B_{X}\to\overline{\mathbb{D}} : \displaystyle(1+\beta)\int_{0}^{1}F_{j}(tz)t^{\beta-1}\;dt=(1+\beta)\sum_{s=m}^{\infty}(P_{s})_{j}(z)\int_{0}^{1}t^{s+\beta-1}\;dt=(1+\beta)\sum_{s=m}^{\infty}\frac{(P_{s})_{j}(z)}{s+\beta}. It is evident that, the majorant series corresponding to the Bernardi operator for a holomorphic mapping F:B_{X}\to\overline{\mathbb{D}^{n}} is \displaystyle(1+\beta)\sum_{s=m}^{\infty}\frac{\|P_{s}(z)\|_{\infty}}{s+\beta}. In the following result, we establish the Bohr-type inequality for the Bernardi operator defined on a set of holomorphic mappings F:B_{X}\to\overline{\mathbb{D}^{n}} . Theorem 2.2 . Let B_{X} be the unit ball of a complex Banach space X , m\in\mathbb{N}\cup\{0\} and \beta>-m . Let F:B_{X}\to\overline{\mathbb{D}^{n}} be a holomorphic mapping with \displaystyle F(z)=P_{m}(z)+\sum_{s=m+1}^{\infty}P_{s}(z),\quad z\in B_{X}, where P_{s}(z)=(1/s!)D^{s}F(0)(z^{s}) . Assume that \|P_{m}(z)\|_{\infty} is constant on \partial B_{X} and |(P_{m})_{j}(z_{0})|=\|P_{m}(z_{0})\|_{\infty} for all j\in\{1,2,\ldots,n\} and z_{0}\in\partial B_{X} . Then, we have \displaystyle\sum_{s=m}^{\infty}\frac{\left\|P_{s}(z)\right\|_{\infty}}{\beta+s}\leq\frac{r^{m}}{m+\beta} (2.6) for \|z\|=r\leq R(\beta) , where R(\beta)\in(0,1) is the unique root of the equation \displaystyle\frac{1}{\beta+m}-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s-m}=0. The number R(\beta) cannot be improved. Proof. Let z_{0}\in\partial B_{X} be fixed. Since F=(F_{1},F_{2},\cdots,F_{n}) , let f_{j}(t)=F_{j}(tz_{0}) for t\in\mathbb{D} . Then, f_{j}\in H(\mathbb{D},\overline{\mathbb{D}}) and \displaystyle f_{j}(t)=t^{m}\left((P_{m})_{j}(z_{0})+\sum_{s=m+1}^{\infty}(P_{s})_{j}(z_{0})t^{s-m}\right),\quad t\in\mathbb{D}. Let g_{j}(t)=(P_{m})_{j}(z_{0})+\sum_{s=m+1}^{\infty}(P_{s})_{j}(z_{0})t^{s-m} for t\in\mathbb{D} . Then, g_{j}(t)\in H(\mathbb{D},\overline{\mathbb{D}}) . In view of Lemma 2.1 , for each fixed j (j=1,2,\cdots,n) , we have \displaystyle|(P_{s})_{j}(z_{0})|\leq 1-|(P_{m})_{j}(z_{0})|^{2}=1-\|P_{m}(z_{0})\|^{2}_{\infty}\quad\text{for}\quad s\geq m+1. Thus, we have \displaystyle\|P_{s}(z_{0})\|_{\infty}\leq 1-\|P_{m}(z_{0})\|^{2}_{\infty}\quad\text{for}\quad s\geq m+1. (2.7) Let x=\|P_{m}(z_{0})\|_{\infty}\in[0,1] . If x=1 , then P_{s}(z_{0})=0 for all s\geq m+1 . Therefore, the inequality ( 2.6 ) holds for z=rz_{0} and r\in[0,1) . We assume that x\in[0,1) . For r\in(0,1) and by using ( 2.7 ), we have \displaystyle\sum_{s=m}^{\infty}\frac{\left\|P_{s}(rz_{0})\right\|_{\infty}}{\beta+s} \displaystyle\leq \displaystyle\frac{\|P_{m}(z_{0})\|_{\infty}}{\beta+m}r^{m}+(1-\|P_{m}(z_{0})\|_{\infty}^{2})\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s} \displaystyle= \displaystyle\frac{x}{\beta+m}r^{m}+(1-x^{2})\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s}:=G_{2}(x,r). Differentiating partially G_{2}(x,r) twice with respect to x , we obtain \displaystyle\frac{\partial}{\partial x}G_{2}(x,r)=\frac{1}{\beta+m}r^{m}-2x\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s} and \displaystyle\frac{\partial^{2}}{\partial x^{2}}G_{2}(x,r)=-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s}\leq 0. Thus, \frac{\partial}{\partial x}G_{2}(x,r) is a monotonically decreasing function of x\in[0,1) and it follows that \displaystyle\frac{\partial}{\partial x}G_{2}(x,r)\geq\lim_{x\to 1^{-}}\frac{\partial}{\partial x}G_{2}(x,r)=\frac{1}{\beta+m}r^{m}-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s}\geq 0 for r\leq R(\beta) , where R(\beta)\in(0,1) is the unique positive root of the equation \displaystyle G_{3}(r):=\frac{1}{\beta+m}-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s-m}=0. It is evident that G_{3}(r) is a monotonically decreasing function of r\in(0,1) and \lim_{r\to 0^{+}}G_{3}(r)=1/(\beta+m)>0 . Therefore, G_{2}(x,r) is a monotonically increasing function of x for r\leq R(\beta) and it follows that \displaystyle\sum_{s=m}^{\infty}\frac{\left\|P_{s}(rz_{0})\right\|_{\infty}}{\beta+s}\leq G_{2}(x,r)\leq\lim_{x\to 1^{-}}G_{2}(x,r)=\frac{1}{m+\beta}r^{m}\quad\text{for}\quad r\leq R(\beta). To prove the sharpness of the constant R(\beta) , we consider the function \displaystyle F_{2}(z)=f(l_{z_{0}}(z))u,\quad z\in B_{X}, where z_{0}\in\partial B_{X} , l_{z_{0}}\in T(z_{0}) , u=(u_{1},u_{2},\dots,u_{n})\in\partial\mathbb{D}^{n} with |u_{1}|=|u_{2}|=\dots=|u_{n}|=1 , and \displaystyle f(t)=t^{m}\frac{\lambda-t}{1-\lambda t},\quad t\in\mathbb{D},~\lambda\in(0,1). Thus, we have \displaystyle F_{2}(rz_{0})=f(l_{z_{0}}(rz_{0}))u=\sum_{s=m}^{\infty}P_{s}(rz_{0}), where P_{m}(rz_{0})=\lambda r^{m}u and P_{s}(rz_{0})=\lambda^{s-1}(\lambda^{2}-1)r^{s}u for s\geq m+1 . Therefore, \displaystyle\sum_{s=m}^{\infty}\frac{\left\|P_{s}(rz_{0})\right\|_{\infty}}{\beta+s} \displaystyle= \displaystyle\frac{\lambda}{\beta+m}r^{m}+(1-\lambda^{2})\sum_{s=m+1}^{\infty}\frac{\lambda^{s-1}}{\beta+s}r^{s} (2.8) \displaystyle= \displaystyle\frac{1}{\beta+m}r^{m}-(1-\lambda)r^{m}\left(\frac{1}{\beta+m}-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s-m}\right)+H_{\lambda}(r), where \displaystyle H_{\lambda}(r)=2(\lambda-1)\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s}+(1-\lambda^{2})\sum_{s=m+1}^{\infty}\frac{\lambda^{s-1}}{\beta+s}r^{s}. Letting \lambda\to 1^{-} , we obtain \displaystyle H_{\lambda}(r)=\sum_{s=m+1}^{\infty}\frac{2(\lambda-1)+(1-\lambda^{2})\lambda^{s-1}}{\beta+s}r^{s}=O((1-\lambda)^{2}). (2.9) Furthermore, for r>R(\beta) , we have the following inequality \displaystyle\frac{1}{\beta+m}-2\sum_{s=m+1}^{\infty}\frac{1}{\beta+s}r^{s-m}<0. (2.10) From ( 2.8 ), ( 2.9 ) and ( 2.10 ), we conclude that R(\beta) cannot be improved. This completes the proof. ∎ In view of ( 1.4 ), we consider the majorant series corresponding to the discrete Fourier transform for a holomorphic mapping F:B_{X}\to\overline{\mathbb{D}^{n}} with F(z)=a+\sum_{s=1}^{\infty}P_{s}(z) , where z\in B_{X} and P_{s}(z)=(1/s!)D^{s}F(0)(z^{s}) as follows: \displaystyle\mathcal{F}_{F}(\|z\|):=\sum_{n=0}^{\infty}\|a\|_{\infty}\|z\|^{n}+\sum_{n=1}^{\infty}\left(\sum_{k=1}^{n}\|z\|^{k-1}\|P_{n-k+1}(z)\|_{\infty}\right). In the following result, we establish the Bohr-type inequality for the discrete Fourier transform defined on a set of holomorphic mappings F:B_{X}\to\overline{\mathbb{D}^{n}} . Theorem 2.3 . Let B_{X} be the unit ball of a complex Banach space X . Let a=(a_{1},a_{2},\ldots,a_{n})\in\overline{\mathbb{D}^{n}} and let F:B_{X}\to\overline{\mathbb{D}^{n}} be a holomorphic mapping with \displaystyle F(z)=a+\sum_{s=1}^{\infty}P_{s}(z),\quad z\in B_{X}, where P_{s}(z)=(1/s!)D^{s}F(0)(z^{s}) . Assume that |a_{j}|=\|a\|_{\infty} for all j\in\{1,2,\ldots,n\} . Then, we have \displaystyle\mathcal{F}_{F}(\|z\|)\leq\frac{1}{1-r}\quad\text{for}\quad\|z\|=r\leq 1/3. (2.11) The number 1/3 cannot be improved. Proof. Let z_{0}\in\partial B_{X} be fixed. Since F=(F_{1},F_{2},\cdots,F_{n}) , we consider f_{j}(t)=F_{j}(tz_{0}) for t\in\mathbb{D} . Then, f_{j}\in H(\mathbb{D},\overline{\mathbb{D}}) and \displaystyle f_{j}(t)=a_{j}+\sum_{s=1}^{\infty}(P_{s})_{j}(z_{0})t^{s},\quad t\in\mathbb{D}. Using similar argument as in the proof of Theorem 2.1 , we have \displaystyle\|P_{s}(z_{0})\|_{\infty}\leq 1-\|a\|^{2}_{\infty}\quad\text{for}\quad s\geq 1. (2.12) Let x=\|a\|_{\infty}\in[0,1] . It is evident that, if x=1 , then P_{s}(z_{0})=0 for all s\geq 1 . Therefore, the inequality ( 2.11 ) holds for z=rz_{0} and r\in[0,1) . Thus, we may assume that x\in[0,1) . Using ( 2.12 ), for r\in(0,1) , we have \displaystyle\sum_{n=0}^{\infty}x\|rz_{0}\|^{n}+\sum_{n=1}^{\infty}\left(\sum_{k=1}^{n}\|rz_{0}\|^{k-1}\|P_{n-k+1}(rz_{0})\|_{\infty}\right) \displaystyle\leq \displaystyle x\sum_{n=0}^{\infty}r^{n}+\sum_{n=1}^{\infty}\left((1-x^{2})\sum_{k=1}^{n}r^{n}\right) \displaystyle= \displaystyle\frac{x}{1-r}+(1-x^{2})\sum_{n=1}^{\infty}nr^{n} \displaystyle= \displaystyle\frac{x}{1-r}+(1-x^{2})\frac{r}{(1-r)^{2}}:=G_{4}(x,r). By partially differentiating G_{4}(x,r) twice with respect to x , we obtain \displaystyle\frac{\partial}{\partial x}G_{4}(x,r)=\frac{1}{1-r}-2x\frac{r}{(1-r)^{2}}\quad\text{and}\quad\frac{\partial^{2}}{\partial x^{2}}G_{4}(x,r)=-2\frac{r}{(1-r)^{2}}\leq 0, which shows that \frac{\partial}{\partial x}G(x,r) is a monotonically decreasing function of x\in[0,1) and it follows that \displaystyle\frac{\partial}{\partial x}G_{4}(x,r)\geq\lim_{x\to 1^{-}}\frac{\partial}{\partial x}G_{4}(x,r)=\frac{1-3r}{(1-r)^{2}}\geq 0 for r\leq 1/3 . Thus, G_{4}(x,r) is a monotonically increasing function of x\in[0,1) for r\leq 1/3 and it follows that \displaystyle G_{4}(x,r)\leq\lim_{x\to 1^{-}}G_{4}(x,r)=\frac{1}{1-r}\quad\text{for}\quad r\leq\frac{1}{3}. To prove the sharpness of the constant 1/3 , we consider the function \displaystyle F_{3}(z)=f(l_{z_{0}}(z))u,\quad z\in B_{X}, where z_{0}\in\partial B_{X} , l_{z_{0}}\in T(z_{0}) , u=(u_{1},u_{2},\dots,u_{n})\in\partial\mathbb{D}^{n} with |u_{1}|=|u_{2}|=\dots=|u_{n}|=1 , and \displaystyle f(t)=\frac{\lambda-t}{1-\lambda t},\quad t\in\mathbb{D},~\lambda\in(0,1). Thus, we have \displaystyle F_{3}(rz_{0})=f(l_{z_{0}}(rz_{0}))u=a+\sum_{s=1}^{\infty}P_{s}(rz_{0}), where a=\lambda u and P_{s}(z_{0})=\lambda^{s-1}(\lambda^{2}-1)u for s\geq 1 . Therefore, \displaystyle\sum_{n=0}^{\infty}\|\lambda u\|_{\infty}\|rz_{0}\|^{n}+\sum_{n=1}^{\infty}\left(\sum_{k=1}^{n}\|rz_{0}\|^{k-1}\|P_{n-k+1}(rz_{0})\|_{\infty}\right) (2.13) \displaystyle= \displaystyle\sum_{n=0}^{\infty}\lambda r^{n}+\sum_{n=1}^{\infty}\left(\sum_{k=1}^{n}r^{n}\lambda^{n-k}(1-\lambda^{2})\right) \displaystyle= \displaystyle\frac{\lambda}{1-r}+(1-\lambda^{2})\sum_{n=1}^{\infty}r^{n}\left(\sum_{k=1}^{n}\lambda^{n-k}\right) \displaystyle= \displaystyle\frac{\lambda}{1-r}+(1-\lambda^{2})\sum_{n=1}^{\infty}r^{n}\cdot\frac{1-\lambda^{n}}{1-\lambda} \displaystyle= \displaystyle\frac{\lambda}{1-r}+(1+\lambda)\left(\frac{r}{1-r}-\frac{\lambda r}{1-\lambda r}\right) \displaystyle= \displaystyle\frac{1}{1-r}-(1-\lambda)\frac{1-3r}{1-r}+G_{5}(\lambda,r), where \displaystyle G_{5}(\lambda,r) \displaystyle= \displaystyle\frac{(1+\lambda)}{(1-r)}\left(r-\frac{\lambda r(1-r)}{1-\lambda r}\right)+(1-\lambda)\frac{1-3r}{1-r}-\frac{1-\lambda}{1-r} \displaystyle= \displaystyle\frac{(1+\lambda)}{(1-r)}\left(\frac{r(1-\lambda)}{1-\lambda r}\right)-(1-\lambda)\frac{3r}{1-r} \displaystyle= \displaystyle(1-\lambda)\frac{r}{1-r}\left(\frac{\lambda+3\lambda r-2}{1-\lambda r}\right)~\text{which tends to}~0~\text{as}~\lambda\to 1^{-}. Furthermore, for r>1/3 , we have (1-\lambda)(1-3r)/(1-r)<0 . From ( 2.13 ), we conclude that 1/3 cannot be improved. This completes the proof. ∎ Using similar argument to those used for establishing the majorant series corresponding to the Cesáro operator for a holomorphic mapping F:B_{X}\to\overline{D}^{n} and in view of ( 1.5 ), we obtain \displaystyle\int_{0}^{1}\frac{F_{j}(tz)}{(1-tz)^{\beta}}dt=\sum_{n=0}^{\infty}\frac{a_{j}\Gamma(n+\beta)}{\Gamma(n+2)\Gamma(\beta)}z^{n}+\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}z^{k-1}(P_{n-k+1})_{j}(z)\right), where F_{j}:B_{X}\to\overline{\mathbb{D}} is a holomorphic mapping such that F_{j}(z)=a_{j}+\sum_{s=1}^{\infty}(P_{s})_{j}(z) , z\in B_{X} and j\in\{1,2\ldots,n\} . Evidently, the majorant series corresponding to the \beta -Cesáro operator for a holomorphic mapping F:B_{X}\to\overline{\mathbb{D}^{n}} is given by \displaystyle\mathcal{C}_{\beta}^{F}(\|z\|):=\sum_{n=0}^{\infty}\frac{\|a\|_{\infty}\Gamma(n+\beta)}{\Gamma(n+2)\Gamma(\beta)}\|z\|^{n}+\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}\|z\|^{k-1}\|P_{n-k+1}(z)\|_{\infty}\right). In the following result, we establish the Bohr-type inequality for the \beta -Cesáro operator defined on a set of holomorphic mappings F:B_{X}\to\overline{\mathbb{D}^{n}} . Theorem 2.4 . Let B_{X} be the unit ball of a complex Banach space X and 0<\beta(\not=1) . Let a=(a_{1},a_{2},\ldots,a_{n})\in\overline{\mathbb{D}^{n}} and let F:B_{X}\to\overline{\mathbb{D}^{n}} be a holomorphic mapping with \displaystyle F(z)=a+\sum_{s=1}^{\infty}P_{s}(z),\quad z\in B_{X}, where P_{s}(z)=(1/s!)D^{s}F(0)(z^{s}) . Assume that |a_{j}|=\|a\|_{\infty} for all j\in\{1,2,\ldots,n\} . Then, we have \displaystyle\mathcal{C}_{\beta}^{F}(\|z\|)\leq\frac{1}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right)\quad\text{for}\quad\|z\|=r\leq R_{\beta}, (2.14) where R_{\beta}\in(0,1) is the positive root of the equation \displaystyle\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}=0. The number R_{\beta} cannot be improved. Proof. Let z_{0}\in\partial B_{X} be fixed. Since F=(F_{1},F_{2},\cdots,F_{n}) , we consider the function f_{j}(t)=F_{j}(tz_{0}) for t\in\mathbb{D} . Then, f_{j}\in H(\mathbb{D},\overline{\mathbb{D}}) and \displaystyle f_{j}(t)=a_{j}+\sum_{s=1}^{\infty}(P_{s})_{j}(z_{0})t^{s},\quad t\in\mathbb{D}. Using similar argument as in the proof of Theorem 2.1 , we have \displaystyle\|P_{s}(z_{0})\|_{\infty}\leq 1-\|a\|^{2}_{\infty}\quad\text{for}\quad s\geq 1. (2.15) Let x=\|a\|_{\infty}\in[0,1] . Evidently, if x=1 , then P_{s}(z_{0})=0 for all s\geq 1 . Therefore, the inequality ( 2.14 ) holds for z=rz_{0} and r\in[0,1) . Let x\in[0,1) . Using ( 1.6 ), ( 1.7 ) and ( 2.15 ), for r\in(0,1) , we have \displaystyle x\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}\|rz_{0}\|^{n} \displaystyle+\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}\|rz_{0}\|^{k-1}\|P_{n-k+1}(rz_{0})\|_{\infty}\right) \displaystyle\leq \displaystyle x\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}r^{n}+(1-x^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}\right) \displaystyle= \displaystyle x\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}r^{n}+(1-x^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\right) \displaystyle= \displaystyle\frac{x}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta}}dt+\frac{1-x^{2}}{r}\int_{0}^{r}\frac{t}{(1-t)^{\beta+1}}dt \displaystyle= \displaystyle\frac{x^{2}+x-1}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta}}dt+\frac{1-x^{2}}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta+1}}dt \displaystyle= \displaystyle\frac{1}{r}\left(\frac{(x^{2}+x-1)\left(1-(1-r)^{1-\beta}\right)}{1-\beta}+\frac{(1-x^{2})\left((1-r)^{-\beta}-1\right)}{\beta}\right):=G_{6}(x,r). By partially differentiating G_{6}(x,r) twice with respect to x , we obtain \displaystyle\frac{\partial}{\partial x}G_{6}(x,r)=\frac{1}{r}\left(\frac{(2x+1)\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2x\left((1-r)^{-\beta}-1\right)}{\beta}\right) and \displaystyle\frac{\partial^{2}}{\partial x^{2}}G_{6}(x,r)=\frac{1}{r}\left(\frac{2\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}\right)\leq 0, which shows that \frac{\partial}{\partial x}G_{6}(x,r) is a monotonically decreasing function of x\in[0,1) and it follows that \displaystyle\frac{\partial}{\partial x}G_{6}(x,r)\geq\lim_{x\to 1^{-}}\frac{\partial}{\partial x}G(x,r)=\frac{1}{r}\left(\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}\right)\geq 0 for r\leq R_{\beta} , where R_{\beta}\in(0,1) is the positive root of the equation \displaystyle\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}=0. Thus, G_{6}(x,r) is a monotonically increasing function of x\in[0,1) for r\leq R_{\beta} and it follows that \displaystyle G_{6}(x,r)\leq\lim_{x\to 1^{-}}G_{6}(x,r)=\frac{1}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right)\quad\text{for}\quad r\leq R_{\beta}. To prove the sharpness of the constant R_{\beta} , we consider the function \displaystyle F_{4}(z)=f(l_{z_{0}}(z))u,\quad z\in B_{X}, where z_{0}\in\partial B_{X} , l_{z_{0}}\in T(z_{0}) , u=(u_{1},u_{2},\dots,u_{n})\in\partial\mathbb{D}^{n} with |u_{1}|=|u_{2}|=\dots=|u_{n}|=1 , and \displaystyle f(t)=\frac{\lambda-t}{1-\lambda t},\quad t\in\mathbb{D},~\lambda\in(0,1). Then, we have F_{4}(rz_{0})=f(l_{z_{0}}(rz_{0}))u=a+\sum_{s=1}^{\infty}P_{s}(rz_{0}) , where a=\lambda u and P_{s}(z_{0})=\lambda^{s-1}(\lambda^{2}-1)u for s\geq 1 . Using ( 1.6 ) and ( 1.7 ), we have \displaystyle\|\lambda u\|_{\infty}\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}\|rz_{0}\|^{n}\hskip 85.35826pt \displaystyle+\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}\|rz_{0}\|^{k-1}\|P_{n-k+1}(rz_{0})\|_{\infty}\right) \displaystyle= \displaystyle\lambda\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}r^{n}+\sum_{n=1}^{\infty}\frac{1}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}r^{n}\lambda^{n-k}(1-\lambda^{2})\right) \displaystyle= \displaystyle\lambda\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}r^{n}+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(k-1+\beta)}{\Gamma(k)\Gamma(\beta)}\lambda^{n-k}\right) \displaystyle= \displaystyle\frac{\lambda}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta}}dt+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k-1}\right) \displaystyle= \displaystyle\frac{\lambda}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right)+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k-1}\right) (2.16) \displaystyle= \displaystyle\frac{1}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right)-\frac{1-\lambda}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right) \displaystyle+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k-1}\right) \displaystyle= \displaystyle\frac{1}{r}\left(\frac{1-(1-r)^{1-\beta}}{1-\beta}\right)-\frac{1-\lambda}{r}\left(\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}\right) \displaystyle+G_{7}(\lambda,r), where \displaystyle G_{7}(\lambda,r) \displaystyle= \displaystyle\frac{1-\lambda}{r}\left(\frac{2\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}\right)\hskip 113.81102pt \displaystyle+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k-1}\right) \displaystyle= \displaystyle 2(1-\lambda)\left(\frac{1}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta}}dt-\frac{1}{r}\int_{0}^{r}\frac{1}{(1-t)^{\beta+1}}dt\right) \displaystyle+(1-\lambda^{2})\sum_{n=1}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=1}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k-1}\right) \displaystyle= \displaystyle 2(1-\lambda)\left(\sum_{n=0}^{\infty}\frac{1}{n+1}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}r^{n}\right. \displaystyle\left.-\sum_{n=0}^{\infty}\frac{1}{n+1}\left(\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\right)r^{n}\right) \displaystyle+\frac{(1-\lambda^{2})}{\lambda}\sum_{n=0}^{\infty}\frac{r^{n}}{n+1}\left(\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k}-\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}\right) \displaystyle= \displaystyle\sum_{n=0}^{\infty}\frac{1}{n+1}\left(-\frac{(1-\lambda)^{2}}{\lambda}\frac{\Gamma(n+\beta)}{\Gamma(n+1)\Gamma(\beta)}-2(1-\lambda)\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\right. \displaystyle\left.+\frac{(1-\lambda^{2})}{\lambda}\sum_{k=0}^{n}\frac{\Gamma(n-k+\beta)}{\Gamma(n-k+1)\Gamma(\beta)}\lambda^{k}\right)r^{n}. Using the identity ( 1.9 ) and letting \lambda\to 1^{-} , we have \displaystyle G_{7}(\lambda,r)=O\left((1-\lambda)^{2}\right). (2.17) Furthermore, it is evident that for r>R_{\beta} , the following inequality holds \displaystyle\frac{3\left(1-(1-r)^{1-\beta}\right)}{1-\beta}-\frac{2\left((1-r)^{-\beta}-1\right)}{\beta}<0. (2.18) From ( 2.16 ), ( 2.17 ) and ( 2.18 ), we conclude that R_{\beta} cannot be improved. This completes the proof. ∎ Declarations Authors’ contributions: All authors have equal contribution to complete the manuscript. All of them read and approved the final manuscript. Conflict of Interest: Authors declare that they have no conflict of interest. Availability of data and materials: Not applicable. Acknowledgment: Second author is supported by University Grants Commission (IN) fellowship (No. F. 44 - 1/2018 (SA - III)).