ON CERTAIN PROPERTIES OF NEIGHBORHOODS OF MULTIVALENT FUNCTIONS INVOLVING THE GENERALIZED SAITOH OPERATOR
HESAM MAHZOON AND S. LATHA DEPARTMENT OFSTUDIES INMATHEMATICS
MANASAGANGOTRIUNIVERSITY OFMYSORE- INDIA.
mahzoon_hesam@yahoo.com DEPARTMENT OFMATHEMATICS
YUVARAJA’SCOLLEGEUNIVERSITY OFMYSORE-INDIA.
drlatha@gmail.com
Received 10 June, 2009; accepted 03 November, 2009 Communicated by S.S. Dragomir
ABSTRACT. In this paper, we introduce the generalized Saitoh operatorLp(a, c, η)and using this operator, the new subclasses Hp,bn,m(a, c, η), Lp,bn,m(a, c, η;µ), Hp,b,αn,m(a, c, η) and Lp,b,αn,m(a, c, η;µ) of the class of multivalent functions denoted byAp(n)are defined. Further for functions belonging to these classes, certain properties of neighborhoods are studied.
Key words and phrases: Coefficient bounds,(n, δ)- neighborhood and generalized Saitoh operator.
2000 Mathematics Subject Classification. 30C45.
1. INTRODUCTION
LetAp(n)be the class of normalized functionsf of the form
(1.1) f(z) =zp+
∞
X
k=n+p
akzk, (n, p∈N),
which are analytic andp-valent in the open unit disc U ={z ∈C : |z|<1}.
LetTp(n)be the subclass ofAp(n),consisting of functionsf of the form
(1.2) f(z) = zp −
∞
X
k=n+p
akzk, (ak≥0, n, p∈N),
which arep-valent inU.
The Hadamard product of two power series f(z) = zp+
∞
X
k=n+p
akzk and g(z) = zp+
∞
X
k=n+p
bkzk
155-09
is defined as
(f ∗g)(z) = zp+
∞
X
k=n+p
akbkzk.
Definition 1.1. Fora ∈R, c ∈ R\Z−0,where Z−0 ={...,−2,−1,0}andη ∈ R(η ≥ 0),the operatorLp(a, c, η) :Ap(n)→ Ap(n), is defined as
(1.3) Lp(a, c, η)f(z) =φp(a, c, z)∗Dηf(z), where
Dηf(z) = (1−η)f(z) + η
pzf0(z), (η≥0, z∈ U) and
φp(a, c, z) =zp+
∞
X
k=n+p
(a)k−p
(c)k−p
zk, z∈ U
and(x)kdenotes the Pochammer symbol given by (x)k=
( 1 if k = 0,
x(x+ 1)· · ·(x+k−1) if k ∈N={1,2,3, ...}.
In particular, we have,L1(a, c, η)≡L(a, c, η).
Further, iff(z) =zp+P∞
k=n+pakzk,then Lp(a, c, η)f(z) =zp+
∞
X
k=n+p
1 +
k p −1
η
(a)k−p
(c)k−p
akzk.
Remark 1. Forη = 0andn= 1,we obtain the Saitoh operator [7] which yields the Carlson - Shaffer operator [1] forη= 0 andn=p= 1.
For any function f ∈ Tp(n)andδ≥0,the(n, δ)-neighborhood of f is defined as, (1.4) Nn,δ(f) =
(
g ∈ Tp(n) :g(z) = zp−
∞
X
k=n+p
bkzk and
∞
X
k=n+p
k|ak−bk| ≤δ )
.
For the function h(z) =zp, (p∈N)we have, (1.5) Nn,δ(h) =
(
g ∈ Tp(n) :g(z) = zp−
∞
X
k=n+p
bkzk and
∞
X
k=n+p
k|bk| ≤δ )
.
The concept of neighborhoods was first introduced by Goodman [2] and then generalized by Ruscheweyh [6] .
Definition 1.2. A functionf ∈ Tp(n)is said to be in the class Hp,bn,m(a, c, η) if (1.6)
1 b
z(Lp(a, c, η)f(z))(m+1)
(Lp(a, c, η)f(z))(m) −(p−m)
!
<1,
where p∈N, m∈N0, a >0, η≥0, p > m, b∈C\ {0} and z ∈ U. Definition 1.3. A functionf ∈ Tp(n)is said to be in the class Lp,bn,m(a, c, η;µ) if (1.7)
1 b
"
p(1−µ)
Lp(a, c, η)f(z) z
(m)
+µ(Lp(a, c, η)f(z))(m+1)−(p−m)
#
< p−m,
where p∈N, m ∈N0, a >0, η≥0, p > m, µ≥0, b∈C\ {0}and z ∈ U.
2. COEFFICIENTBOUNDS
In this section, we determine the coefficient inequalities for functions to be in the subclasses Hp,bn,m(a, c, η)andLp,bn,m(a, c, η;µ).
Theorem 2.1. Letf ∈ Tp(n). Then, f ∈ Hp,bn,m(a, c, η) if and only if
(2.1)
∞
X
k=n+p
1 +
k p −1
η
(a)k−p (c)k−p
k m
(k+|b| −p)ak ≤ |b|
p m
.
Proof. Let f ∈ Hp,bn,m(a, c, η).Then, by(1.6)and(1.7)we can write, (2.2) <
P∞
k=n+p
h 1 +
k p −1
ηi(a)
k−p
(c)k−p
k m
(p−k)akzk−p
p m
−P∞ k=n+p
h 1 +
k p −1
ηi(a)
k−p
(c)k−p
k m
akzk−p
>−|b|, (z ∈ U).
Takingz = r, (0 ≤ r <1)in(2.2),we see that the expression in the denominator on the left hand side of(2.2), is positive forr = 0 and for allr, 0 ≤ r < 1.Hence, by letting r 7→ 1− through real values, expression(2.2)yields the desired assertion(2.1).
Conversely, by applying the hypothesis(2.1)and letting|z|= 1,we obtain,
z(Lp(a, c, η)f(z))(m+1)
(Lp(a, c, η)f(z))(m) −(p−m)
=
P∞ k=n+p
h 1 +
k p −1
ηi(a)
k−p
(c)k−p
k m
(p−k)akzk−m
p m
zp−m−P∞ k=n+p
h 1 +
k p −1
ηi(a)
k−p
(c)k−p
k m
akzk−m
≤
|b|h
p m
−P∞ k=n+p
h 1 +
k p −1
ηi(a)
k−p
(c)k−p
k m
aki
p m
−P∞ k=n+p
h 1 +
k p −1
η
i(a)
k−p
(c)k−p
k m
ak
=|b|.
Hence, by the maximum modulus theorem, we have f ∈ Hp,bn,m(a, c, η).
On similar lines, we can prove the following theorem.
Theorem 2.2. A function f ∈ Lp,bn,m(a, c, η;µ) if and only if
(2.3)
∞
X
k=n+p
1 +
k p −1
η
(a)k−p
(c)k−p
k−1 m
[µ(k−1) + 1]ak
≤(p−m)
|b| −1
m! +
p m
.
3. INCLUSIONRELATIONSHIPS INVOLVING(n, δ)-NEIGHBORHOODS
In this section, we prove certain inclusion relationships for functions belonging to the classes Hp,bn,m(a, c, η)andLp,bn,m(a, c, η;µ).
Theorem 3.1. If
(3.1) δ = (n+p)|b| mp
(n+|b|)
1 + npη
(a)n
(c)n
n+p m
, (p > |b|),
then Hn,mp,b (a, c, η)⊂ Nn,δ(h).
Proof. Let f ∈ Hp,bn,m(a, c, η). By Theorem 2.1, we have, (n+|b|)
1 + n
pη (a)n
(c)n
n+p m
∞ X
k=n+p
ak ≤ |b|
p m
,
which implies (3.2)
∞
X
k=n+p
ak≤ |b| mp
(n+|b|)
1 + npη
(a)n
(c)n
n+p m
. Using(2.1)and(3.2), we have,
1 + n
pη (a)n
(c)n
n+p m
∞ X
k=n+p
kak
≤ |b|
p m
+ (p− |b|)
1 + n pη
(a)n
(c)n
n+p m
∞ X
k=n+p
ak
≤ |b|
p m
+ (p− |b|)
1 + n pη
(a)n (c)n
n+p m
|b| mp (n+|b|)
1 + npη(a)
n
(c)n
n+p m
=|b|
p m
n+p n+|b|. That is,
∞
X
k=n+p
kak ≤ |b|(n+p) mp (n+|b|)
1 + npη
(a)n
(c)n
n+p m
=δ, (p > |b|).
Thus, by the definition given by(1.5), f ∈ Nn,δ(h).
Similarly, we prove the following theorem.
Theorem 3.2. If
(3.3) δ=
(p−m)(n+p)h|b|−1
m! + mpi [µ(n+p−1) + 1]
1 + npη(a)
n
(c)n
n+p m
, (µ >1) thenLp,bn,m(a, c, η;µ)⊂ Nn,δ(h).
4. FURTHER NEIGHBORHOODPROPERTIES
In this section, we determine the neighborhood properties of functions belonging to the sub- classesHn,mp,b,α(a, c, η) andLp,b,αn,m (a, c, η;µ).
For 0 ≤ α < p and z ∈ U, a function f is said to be in the class Hp,b,αn,m (a, c, η) if there exists a function g ∈ Hp,bn,m(a, c, η) such that
(4.1)
f(z) g(z) −1
< p−α.
For 0 ≤ α < p and z ∈ U, a function f is said to be in the classLp,b,αn,m (a, c, η;µ) if there exists a function g ∈ Lp,bn,m(a, c, η;µ) such that the inequality(4.1)holds true.
Theorem 4.1. If g ∈ Hp,bn,m(a, c, η) and
(4.2) α =p−
δ(n+|b|)
1 + npη(a)
n
(c)n
n+p m
(n+p)h
(n+|b|)
1 + npη
(a)n
(c)n
n+p m
− |b| mpi, then Nn,δ(g)⊂ Hp,b,αn,m (a, c, η).
Proof. Let f ∈ Nn,δ(g). Then, (4.3)
∞
X
k=n+p
k|ak−bk| ≤δ,
which yields the coefficient inequality, (4.4)
∞
X
k=n+p
|ak−bk| ≤ δ
n+p, (n∈N).
Since g ∈ Hn,mp,b (a, c, η), by(3.2)we have, (4.5)
∞
X
k=n+p
bk ≤ |b| mp
(n+|b|) 1 + npη
(a)n
(c)n
n+p m
so that,
f(z) g(z) −1
<
P∞
k=n+p|ak−bk| 1−P∞
k=n+pbk
≤ δ n+p
(n+|b|)
1 + npη
(a)n
(c)n
n+p m
h
(n+|b|)
1 + npη
(a)n
(c)n
n+p m
− |b| mpi
=p−α.
Thus, by definition, f ∈ Hp,b,αn,m(a, c, η) forαgiven by(4.2).
On similar lines, we prove the following theorem.
Theorem 4.2. If g ∈ Lp,bn,m(a, c, η;µ) and (4.6)
α =p−
δ[µ(n+p−1) + 1]
1 + npη
(a)n
(c)n
n+p−1 m
(n+p)h
{µ(n+p−1) + 1}
1 + npη
(a)n
(c)n
n+p−1 m
−(p−m)|b|−1
m! + mpi, thenNn,δ(g)⊂ Lp,b,αn,m (a, c, η;µ).
REFERENCES
[1] B.C. CARLSONANDD.B. SHAFFER, Starlike and prestarlike hypergeometric functions, SIAM. J.
Math. Anal.,15(4) (1984), 737–745.
[2] A.W. GOODMAN, Univalent functions and non-analytic curves, Proc. Amer. Math. Soc.,8 (1957), 598–601.
[3] G. MURUGUSUNDARAMOORTHYANDH.M. SRIVASTAVA, Neighborhoods of certain classes of analytic functions of complex order, J. Inequal. Pure and Appl. Math., 5(2) (2004), Art. 24.
[ONLINE:http://jipam.vu.edu.au/article.php?sid=374]
[4] M.A. NASRANDM.K. AOUF, Starlike function of complex order, J. Natur. Sci. Math., 25 (1985), 1–12.
[5] R.K. RAINA AND H.M. SRIVASTAVA, Inclusion and neighborhood properties of some analytic and multivalent functions, J. Inequal. Pure and Appl. Math.,7(1) (2006), Art. 5. [ONLINE:http:
//jipam.vu.edu.au/article.php?sid=640]
[6] S. RUSCHEWEYH, Neighbohoods of univalent functions, Proc. Amer. Math. Soc., 81 (1981), 521–
527.
[7] H. SAITOH, A linear operator and its applications of first order differential subordinations, Math.
Japan, 44 (1996), 31–38.
[8] H.M. SRIVASTAVA AND S. OWA, Current Topics in Analytic Function Theory, World Scientific Publishing Company, 1992.
[9] P. WIATROWSKI, On the coefficients of some family of holomorphic functions, Zeszyty Nauk.
Uniw. Lodz Nauk. Mat. - Przyrod., (2), 39 (1970), 75–85.