Continue from this paper
Indexing is not verification. Extracted statements should be checked against the source.
Abstract
For the real number $α>1$, we use a technique due to Nehari and Netanyahu and an application of certain integral iteration of Caratheodory functions to find the best-possible upper bounds on the coefficients of functions of the class $T_n^α(β)$ introduced in \cite{TOO} by Opoola.
Results & Lemmas (7)
Each result is auto-extracted from the paper via OCR - the math is shown verbatim and may carry occasional transcription artifacts. Click to expand the full statement.
Theorem 1
Theorem 1. ([2]). Let. Define where for each is defined by the power series Also let Then The inequalities are sharp. Equalities are…
Theorem 1. ([2]). Let
$$f \in T_n^{\alpha}(\beta)$$
. Define
$$B_{m} = \frac{2^{m}(1-\beta)^{m}\alpha^{m(n-1)}\prod_{j=0}^{m-1}(1-j\alpha)}{m!}$$
$$A_{k}(n,\alpha,\beta) = \sum_{m=1}^{k-1}B_{m}Q_{k-1}^{(m)}$$
$$A_{k-1}(n,\alpha,\beta) = \sum_{m=1}^{k-2}B_{m}Q_{k-1}^{(m)}$$
where for each $m = 1, 2, ..., Q_{k-1}^{(m)}$ is defined by the power series
$$\left(\sum_{k=1}^{\infty} \frac{z^k}{(\alpha+k)^n}\right)^m = Q_m^{(m)} z^m + Q_{m+1}^{(m)} z^{m+1} + Q_{m+2}^{(m)} z^{m+2} + \dots$$
Also let
$$\Omega_1 = \{\alpha | 0 < \alpha < (k-2)^{-1}, k = 2, 3, ...\},$$
$$\Omega_2 = \{\alpha | (k-2)^{-1} \le \alpha \le (k-3)^{-1}, k = 4, 6, ...\},$$
$$\Omega_3 = \{\alpha | (k-2)^{-1} \le \alpha < (k-3)^{-1}, k = 3, 5, ...\}$$
Then
$$|a_k| \le \begin{cases} A_k & \text{if } \alpha \in \Omega_1 \cup \Omega_2, \\ A_{k-1} & \text{if } \alpha \in \Omega_3. \end{cases}$$
The inequalities are sharp. Equalities are attained for f(z) satisfying
$$\frac{D^n f(z)^{\alpha}}{\alpha^n z^{\alpha}} = \begin{cases} \frac{1 + (1 - 2\beta)z}{1 - z} & \text{if } \alpha \in \Omega_1 \cup \Omega_2, \\ \frac{1 + (1 - 2\beta)z^2}{1 - z^2} & \text{if } \alpha \in \Omega_3. \end{cases}$$
The rough estimate was also given as:
Theorem 2 · coeff
Theorem 2. ([2]). Let. Suppose Then From Theorem 1, which is best-possible, it is obvious that the problem has only been solved completely…
Theorem 2. ([2]). Let $f \in T_n^{\alpha}(\beta), n \in N$ . Suppose
$$\frac{f(z)^{\alpha}}{z^{\alpha}} = \sum_{k=0}^{\infty} A_{k+1}(\alpha)z^k, A_1(\alpha) = 1.$$
Then
$$|A_{k+1}(\alpha)| < \exp\left\{0.624\alpha^2 + (2\alpha^2 - \frac{1}{2})\sum_{j=1}^k \frac{1}{j}\right\}.$$
From Theorem 1, which is best-possible, it is obvious that the problem has only been solved completely for $a_2$ and $a_3$ for all values of the index $\alpha$ . For $k \geq 4$ , the problem has remained open for all $\alpha > (k-3)^{-1}$ . In this article we proceed with the proof of the sharp bounds on the coefficients of functions in the class $T_n^{\alpha}(\beta)$ , $\alpha > 1$ . Our result is the following:
Theorem 3
Theorem 3. Let. If, we have the sharp inequalities (3) Equalities are attained for f(z) satisfying For k=2, 3, the above result is…
Theorem 3. Let $f \in T_n^{\alpha}(\beta)$ . If $\alpha > 1$ , we have the sharp inequalities
(3)
$$|a_k| \le \frac{2(1-\beta)\alpha^{n-1}}{(\alpha+k-1)^n}, \quad k=2, 3, 4, \dots$$
Equalities are attained for f(z) satisfying
$$\frac{D^n f(z)^{\alpha}}{\alpha^n z^{\alpha}} = \frac{1 + (1 - 2\beta)z^{k-1}}{1 - z^{k-1}}.$$
For k=2, 3, the above result is contained in Theorem 1 above. The proof here is however new. Also Singh [6] proved the same result for k=2, 3, 4 for the particular case n=1 and $\beta=0$ . In our proof we combine a method of classical analysis due to Nehari and Netanyau [3] (also used by Singh [6]) with an application of certain integral iteration of the Caratheodory
functions [1]. That is presented in Section 3. In the next section we state and prove some preliminary lemmas.
Lemma 1
Lemma 1. ([3]). If and belong to P, then also belongs to P.
Lemma 1. ([3]). If $p(z) = 1 + \sum_{k=1}^{\infty} b_k z^k$ and $q(z) = 1 + \sum_{k=1}^{\infty} c_k z^k$ belong to P, then $r(z) = 1 + \frac{1}{2} \sum_{k=1}^{\infty} b_k c_k z^k$ also belongs to P.
Lemma 2
Lemma 2. ([3]). Let and be functions in P. Set (6) If is defined by then If, in the proof of the above lemma (as contained in [3]),we…
Lemma 2. ([3]). Let $h(z) = 1 + \sum_{k=1}^{\infty} d_k z^k$ and $1 + G(z) = 1 + \sum_{k=1}^{\infty} b'_k z^k$ be functions in P. Set
(6)
$$\gamma_m = \frac{1}{2^m} \left[ 1 + \frac{1}{2} \sum_{\mu=1}^m {m \choose \mu} d_\mu \right], \quad \gamma_0 = 1.$$
If $A_k$ is defined by
$$\sum_{m=1}^{\infty} (-1)^{m+1} \gamma_{m-1} G_1^m(z) = \sum_{k=1}^{\infty} A_k z^k,$$
then
$$|A_k| < 2, \quad k = 1, 2, \dots$$
If, in the proof of the above lemma (as contained in [3]),we define $h_n(z)$ as the nth iterated integral transform of $h_0(z) = h(z)$ we immediately obtain the following corollary.
Corollary 1
Corollary 1. Let be the nth integral iteration of with Re, and be functions in P. Define as in (6) and (7) If is defined by (8) then (9)
Corollary 1. Let $h_n(z)$ be the nth integral iteration of $h_0(z) = 1 + \sum_{k=1}^{\infty} d_k z^k$ with Re $h_n(z) > \beta$ , and $1 + G(z) = 1 + \sum_{k=1}^{\infty} b'_k z^k$ be functions in P. Define $\gamma_m$ as in (6) and
(7)
$$\eta_m = \frac{(1-\beta)\alpha^n}{(\alpha+m)^n} \gamma_m, \quad \eta_0 = 1-\beta.$$
If $A_k$ is defined by
(8)
$$\sum_{m=1}^{\infty} (-1)^{m+1} \eta_{m-1} G_1^m(z) = \sum_{k=1}^{\infty} A_k z^k,$$
then
(9)
$$|A_k| \le \frac{2(1-\beta)\alpha^n}{(\alpha+k)^n}, \quad k=1, 2, \dots$$
Lemma 3
Lemma 3. ([2]). Let be a power series. Then the mth integer product of G(z) is where and,. We now turn to the proof of the main result.
Lemma 3. ([2]). Let $G(z) = \sum_{k=0}^{\infty} c_k z^k$ be a power series. Then the mth integer product of G(z) is $G^m(z) = \sum_{k=0}^{\infty} c_k^{(m)} z^k$ where $c_k^{(1)} = c_k$ and $c_k^{(m)} = \sum_{j=0}^{k} c_j c_{k-j}^{(m-1)}$ , $m \ge 2$ .
We now turn to the proof of the main result.
Coefficient bounds & claims (2)
Machine-extracted from the paper text - useful for cross-referencing, not a verified fact.
coefficient_bound
|a_k| ≤ 2*(1-beta)*alpha**(n-1) / (alpha+k-1)**n for class T^alpha_n(beta) (sharp) [Theorem 3]
function_family
Class T^alpha_n(beta): f in A satisfying Re(D^n f(z)^alpha / (alpha^n z^alpha)) > beta, where D^n is the Salagean derivative, alpha > 0 real, 0 <= beta < 1
Related Papers