JAM Journal of Applied Mathematics 1687-0042 1110-757X Hindawi Publishing Corporation 127615 10.1155/2013/127615 127615 Research Article Fekete-Szegő Inequality for a Subclass of p-Valent Analytic Functions Raza Mohsan 1 Arif Muhammad 2 Darus Maslina 3 Abd-El-Malek Mina 1 Department of Mathematics Government College University Faisalabad Faisalabad, Punjab 38000 Pakistan gcuf.edu.pk 2 Department of Mathematics Abdul Wali Khan University Mardan Mardan, Khyber Pakhtunkhwa 23200 Pakistan awkum.edu.pk 3 School of Mathematical Sciences Faculty of Science and Technology Universiti Kebangsaan Malaysia 43600 Bangi, Selangor D. Ehsan Malaysia ukm.my 2013 12 6 2013 2013 18 01 2013 11 04 2013 2013 Copyright © 2013 Mohsan Raza et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The main object of this paper is to study Fekete-Szegő problem for the class of p-valent functions. Fekete-Szegő inequality of several classes is obtained as special cases from our results. Applications of the results are also obtained on the class defined by convolution.

1. Introduction and Preliminaries

Let Ap denote the class of functions f(z) of the form (1)f(z)=zp+n=p+1anzn,(p={1,2,3,}), which are analytic in the open unit disk E. Also, A1=A, the usual class of analytic functions defined in the open unit disk E={z:|z|<1}. Let f(z) and g(z) be analytic in E. We say that the function f is subordinate to the function g and write f(z)g(z), if and only if there exists Schwarz function w, analytic in E such that w(0)=0, |w(z)|<1 for zE, and f(z)=g(w(z)). In particular, if g is univalent in E, then we have the following equivalence: (2)f(z)g(z)f(0)=g(0),f(E)g(E).

For any two analytic functions f(z) of the form (1) and g(z) with (3)g(z)=zp+n=p+1bnzn,zE, the convolution (Hadamard product) is given by (4)(f*g)(z)=zp+n=p+1anbnzn,zE. Let ϕ(z) be an analytic function with positive real part on E with ϕ(0)=1, ϕ(0)>0, which maps the unit disk E onto a region starlike with respect to 1 which is symmetric with respect to the real axis. Denote by Sp*(ϕ) the class of functions f analytic in E for which (5)zf(z)pf(z)ϕ(z),zE. The class    Sp*(ϕ) was defined and studied by Ali et al. . They obtained the Fekete-Szegő inequality for functions in the class Sp*(ϕ). The class S1*(ϕ) coincides with the class S*(ϕ) discussed by Ma and Minda . Owa  introduced a subclass of p-valently Bazilevic functions Hp(A,B,α,β). A function fAp is said to be in the class Hp(A,B,α,β) if and only if (6)(1-β)(f(z)zp)α+βzf(z)pf(z)(f(z)zp)α1+Az1+Bz,zE, where -1B<A1, α0, and 0β1. We now define the following subclass of analytic functions.

Definition 1.

Let ϕ(z) be a univalent starlike function with respect to 1 which maps the unit disk E onto a region in the right half-plane which is symmetric with respect to the real axis with ϕ(0)=1 and ϕ(0)>0. A function fAp is in the class Vp,b,α,β(ϕ) if (7)1-2b+2b{zf(z)pf(z)(f(z)zp)α(1-β)(f(z)zp)α1-2b+2b+βzf(z)pf(z)(f(z)zp)α}ϕ(z), where 0β1, α0, and b>0.

Definition 2.

A function fAp is in the class Vp,b,α,β,g(ϕ) if (8)1-2b+2b{βz(f*g)(z)p(f*g)(z)((f*g)(z)zp)α(1-β)((f*g)(z)zp)α1-2b+2b+βz(f*g)(z)p(f*g)(z)((f*g)(z)zp)α}ϕ(z), where 0β1, α0, and b>0.    In other words, a function fAp    is in the class Vp,b,α,β,g(ϕ) if (f*g)(z)Vp,b,α,β(ϕ).

We have the following special cases.

Vp,2,1,1(ϕ) coincides with the class Sp*(ϕ) introduced and studied by Ali et al. .

For p=1, b=2, and β=1, we have the class Bα(ϕ) introduced and studied by Ravichandran et al. .

For b=2 and ϕ(z)=(1+Az)/(1+Bz), the class Vp,2,α,β(ϕ) reduces to Hp(A,B,α,β) introduced and studied by Owa .

For ϕ(z)=(1+(1-2γ)z)/(1-z), the class Vp,2,α,1(ϕ) reduces to the class Bp(α,γ)    defined as (9)Bp(α,γ)={fAp:Re(zf(z)pf(z)(f(z)zp)α)>γ,0γ<1,zE(zf(z)pf(z)(f(z)zp)α)}.

For ϕ(z)=(1+(1-2γ)z)/(1-z), the class Vp,2,α,0(ϕ) is defined as (10){fAp:Re(f(z)zp)α>γ,  0γ<1,  zE}.

V1,2,0,1(ϕ)=S*(ϕ) is investigated by Ma and Minda .

For ϕ(z)=(1+(1-2γ)z)/(1-z), the class V1,2,1,0(ϕ) reduces to the class (11)Bγ={fA1:Ref(z)z>γ}.

studied by Chen .

We need the following results to obtain our main results.

Lemma 3 (see [<xref ref-type="bibr" rid="B1">1</xref>]).

Let Ω be the class of analytic functions w,   normalized by w(0)=0,  satisfying condition |w(z)|<1.    If wΩ    and w(z)=w1z+w2z2+, zE,  then (12)|w2-tw12|{-t,t<-1,1,-1t1,t,t>1. For t-1 or t1,  the equality holds, if and only if w(z)=z or one of its rotation. For -1t1,  the equality holds, if w(z)=z2 or one of its rotation. the equality holds for     t=-1, if and only if w(z)=z((λ+z)/(1+λz))(0λ1) or one of its rotation, while for t=1, the equality holds, if and only if w(z)=-z((λ+z)/(1+λz))(0λ1) or one of its rotation. The above upper bound for -1<t<1 is sharp, and it can be improved as follows: (13)|w2-tw12|+(1+t)|w1|21,-1<t0,|w2-tw12|+(1-t)|w1|21,  0<t<1.

Lemma 4 (see [<xref ref-type="bibr" rid="B4">6</xref>, (7), page 10]).

If wΩ    and w(z)=w1z+w2z2+, zE,  then (14)|w2-tw12|max{1;|t|}, for any complex number t. The result is sharp for the functions w(z)=z2 or w(z)=z.

Lemma 5 (see [<xref ref-type="bibr" rid="B6">7</xref>]).

If wΩ, then for any real number q1 and q2, the following sharp estimate holds: (15)|w3+q1w1w2+q2w13|H(q1,q2), where (16)H(q1,q2)={1,(q1,q2)D1D2,|q2|,(q1,q2)k=37Dk,23(|q1|+1)(|q1|+13(|q1|+1+q2))1/2,(q1,q2)D8D9,13q2(q12-4q12-4q2)(q12-43(q2-1))1/2,(q1,q2)D10D11{±2,1},23(|q1|-1)(|q1|-13(|q1|-1-q2))1/2,(q1,q2)D12. The extremal function up to the rotations is of the form (17)w(z)=z3,w(z)=z,w(z)=w0(z)=z([(1-λ)ε2+λε1]-ε1ε2z)1-[(1-λ)ε1+λε2]z,w(z)=w1(z)=z(t1-z)1-t1z,w(z)=w2(z)=z(t2+z)1+t2z,|ε1|=|ε2|=1,  ε1=t0-e-iθ0/2(ab),ε2=-e-iθ0/2(ia±b),a=t0cosθ02,b=1-t02sin2θ02,λ=b±a2b,t0=(2q2(q12+2)-3q123(q2-1)(q12-4q2))1/2,t1=(|q1|+13(|q1|+1+q2))1/2,t2=(|q1|-13(|q1|-1-q2))1/2,cosθ02=q12(q2(q12+8)-2(q12+2)2q2(q12+2)-3q12). The sets Dk, k=1,2,,12 are defined as follows: (18)D1={(q1,q2):|q1|12,  |q2|1},D2={(q1,q2):12|q1|2,427(|q1|+1)3-(|q1|+1)q21},D3={(q1,q2):|q1|12,  |q2|-1},D4={(q1,q2):|q1|12,  q2-23(|q1|+1)},D5={(q1,q2):|q1|2,  q21},D6={(q1,q2):2|q1|4,  q2112(q12+8)},D7={(q1,q2):|q1|4,  q223(|q1|-1)},D8={(q1,q2):12|q1|2,D8-23(|q1|+1)q2427(|q1|+1)3-(|q1|+1)},D9={2|q1|(|q1|+1)q12+2|q1|+4(q1,q2):|q1|2,-23(|q1|+1)q22|q1|(|q1|+1)q12+2|q1|+4},D10={2|q1|(|q1|+1)q12+2|q1|+4(q1,q2):2|q1|4,2|q1|(|q1|+1)q12+2|q1|+4q2112(q12+8)},D11={2|q1|(|q1|+1)q12+2|q1|+4(q1,q2):|q1|4,2|q1|(|q1|+1)q12+2|q1|+4q22|q1|(|q1|-1)q12-2|q1|+4},D12={2|q1|(|q1|-1)q12-2|q1|+4(q1,q2):|q1|4,2|q1|(|q1|-1)q12-2|q1|+4q223(|q1|-1)}.

2. Main Results Theorem 6.

Let ϕ(z)=1+B1z+B2z2+B3z3+, where Bn, s are real with B1>0 and B20. Let (19)σ1=(αp+β)2bpB12(αp+2β)×{2(B2-B1)-bpB12(α-1)(αp+2β)2(αp+β)2},σ2=(αp+β)2bpB12(αp+2β)×{2(B2+B1)-bpB12(α-1)(αp+2β)2(αp+β)2},σ3=(αp+β)2bpB12(αp+2β)×{2B2-bpB12(α-1)(αp+2β)2(αp+β)2},Φ(p,α,β,μ)=(αp+2β)(2μ+α-1)2(αp+β)2. If f(z) is of the form (1) and belongs to the class Vp,b,α,β(ϕ), then (20)|ap+2-μap+12|{bp2(αp+2β)(B2-bpB122Φ(p,α,β,μ)),  μ<σ1,bpB12(αp+2β),σ1μσ2,-bp2(αp+2β)(B2-bpB122Φ(p,α,β,μ)),μ>σ2. Furthermore, for σ1μσ3, (21)|ap+2-μap+12|+1bpB1(2(1-B2B1)(αp+β)2(αp+2β)+1bpB11bpB1+bpB12(2μ+α-1)(αp+β)2(αp+2β))|ap+1|2bpB12(αp+2β), and for σ3μσ2, (22)|ap+2-μap+12|+1bpB1(2(1+B2B1)(αp+β)2(αp+2β)bpB12(αp+2β)-bpB12(2μ+α-1)(αp+β)2(αp+2β))|ap+1|2bpB12(αp+2β). For any complex number μ, (23)|ap+2-μap+12|bpB12(αp+2β)max{1,|bpB12Φ(p,α,β,μ)-B2B1|}. Also, (24)|ap+3|bpB12(αp+3β)H(q1,q2), where H(q1,q2) is defined in Lemma 3 and (25)q1=2B2B1+bpB12(1-α)(αp+3β)(αp+β)(αp+2β),q2=B3B1+(bpB12)2(α-1)(2α-1)(αp+3β)6(αp+β)3+bpB22(1-α)(αp+3β)(αp+β)(αp+2β). These results are sharp.

Proof.

Since fVp,b,α,β(ϕ), therefore we have for a Schwarz function (26)w(z)=w1z+w2z2+,  zE such that (27)1-2b+2b{(1-β)(f(z)zp)α+βzf(z)pf(z)(f(z)zp)α}=ϕ(w(z)). Now, (28)1-2b+2b{(1-β)(f(z)zp)α+βzf(z)pf(z)(f(z)zp)α}=1+2bp(αp+β)ap+1z+1bp(αp+2β)×{2ap+2+(α-1)ap+12}z2+2(αp+3β)bp×{(α-1)(α-2)6ap+3+(α-1)ap+1ap+2-2b+2b+(α-1)(α-2)6ap+13}z3+. Also, we have (29)ϕ(w(z))=1+B1w1z+(B1w2+B1w12)z2+(B1w3+2B2w1w2+B3w13)z3+. Comparing the coefficients of z, z2, z3 and after simple calculations, we obtain (30)ap+1=bpB1w12(αp+β),ap+2=bpB12(αp+2β)×{w2-(bpB12(α-1)(αp+2β)2(αp+β)2-B2B1)w12},ap+3=bpB12(αp+2β){w3+q1w1w2+q2w13}, where q1 and q2 are defined in (25). It can be easily followed from (30) that (31)ap+2-μap+12=bpB12(αp+2β){w2-νw12}, where (32)ν=bpB12Φ(p,α,β,μ)-B2B1. The results from (20) to (22)    are obtained by using Lemma 3, (23)    by using Lemma 4, and (24) by using Lemma 5. To show that these results are sharp, we define the functions Kϕn(z), Fλ(z), and Gλ(z) such that (33)Kϕn(0)=[Kϕn](0)-1=0,Fλ(0)=Fλ(0)-1=0,Gλ(0)=Gλ(0)-1=0, with (34)1-2b+2b{(1-β)(Kϕn(z)zp)α+βzKϕn(z)pKϕn(z)(Kϕn(z)zp)α}ϕ(zn-1),1-2b+2b{(1-β)(Fλ(z)zp)α+βzFλ(z)pFλ(z)(Fλ(z)zp)α}ϕ(z(z+λ)1+λz),1-2b+2b{(1-β)(Gλ(z)zp)α+βzGλ(z)pGλ(z)(Gλ(z)zp)α}ϕ(-z(z+λ)1+λz). It is clear that the functions Kϕn,Fλ,GλVp,b,α,β(ϕ). Let Kϕ:  =Kϕ2. If μ<σ1 or μ>σ2, then the equality occurs for the function Kϕ or one of its rotations. For σ1<μ<σ2, the equality is attained, if and only if f is Kϕ3 or one of its rotations. When μ=σ1, then the equality holds for the function Fλ or one of its rotations. If μ=σ2, then the equality is obtained for the function Gλ or one of its rotations.

Corollary 7.

For b=2, the results from (20) to (24) coincide with the results proved by Ramachandran et al. .

Corollary 8.

For b=2,  p=1, and β=1,  the results from (20) to (22) coincide with the results obtained by Ravichandran et al.  for the class Bα(ϕ).

Corollary 9.

For b=2,  α=0, and β=1,  the results from (20) to (24) coincide with the results obtained by Ali et al.  for the class Sp*(ϕ).

Corollary 10.

For b=2, p=1, α=0, and β=1,  the results from (20) to (22) coincide with the results obtained by Ma and Minda  for the class S*(ϕ).

2.1. Application of Theorem <xref ref-type="statement" rid="thm2.1">6</xref> to the Function Defined by Convolutions Theorem 11.

Let ϕ(z)=1+B1z+B2z2+B3z3+, where Bn, s are real with B1>0 and B20. Let (35)σ1=gp+12gp+2(αp+β)2bpB12(αp+2β)×{(α-1)(αp+2β)2(αp+β)22(B2-B1)gp+12gp+2-bpB12(α-1)(αp+2β)2(αp+β)2},σ2=gp+12gp+2(αp+β)2bpB12(αp+2β)×{(α-1)(αp+2β)2(αp+β)22(B2+B1)gp+12gp+2-bpB12(α-1)(αp+2β)2(αp+β)2},σ3=gp+12gp+2(αp+β)2bpB12(αp+2β)×{2B2-bpB12(α-1)(αp+2β)2(αp+β)2},Φ*(p,α,β,μ)=(αp+2β)(2μ(gp+2/gp+1)+α-1)2(αp+β)2. If f(z) is of the form (1) and belongs to the class Vp,b,α,β,g(ϕ), then (36)|ap+2-μap+12|{bp2gp+2(αp+2β)(B2-bpB122Φ*(p,α,β,μ)),                                                                                              μ<σ1,bpB12gp+2(αp+2β),                                      σ1μσ2,bp2gp+2(αp+2β)(bpB122Φ*(p,α,β,μ)-B2),                                                                                                    μ>σ2. Furthermore, for σ1μσ3, (37)|ap+2-μap+12|+gp+12bpB1gp+2(2(1-B2B1)(αp+β)2(αp+2β)+gp+12bpB1gp+2B2B1+bpB12(2μgp+2gp+1+α-1)(αp+β)2(αp+2β))|ap+1|2bpB12gp+2(αp+2β), and for σ3μσ2, (38)|ap+2-μap+12|+gp+12bpB1gp+2(2(1+B2B1)(αp+β)2(αp+2β)+gp+12bpB1gp+2(1+B2B1)-bpB12(2μgp+2gp+1+α-1)(αp+β)2(αp+2β))|ap+1|2bpB12gp+2(αp+2β). For any complex number μ, (39)|ap+2-μap+12|bpB12gp+2(αp+2β)max{1,|bpB12Φ*(p,α,β,μ)-B2B1|}. Also, (40)|ap+3|bpB12(αp+3β)gp+3H(q1,q2), where H(q1,q2)    is defined in Lemma 5 and (41)q1=2B2B1+bpB12(1-α)(αp+3β)(αp+β)(αp+2β),q2=B3B1+(bpB12)2(α-1)(2α-1)(αp+3β)6(αp+β)3+bpB22(1-α)(αp+3β)(αp+β)(αp+2β). These results are sharp.

Proof.

Since fVp,b,α,β,g(ϕ), therefore we have for a Schwarz function w,  such that (42)1-2b+2b{βz(f*g)(z)p(f*g)(z)(1-β)((f*g)(z)zp)α1-2b+2b+βz(f*g)(z)p(f*g)(z)((f*g)(z)zp)α}=ϕ(w(z)). Now, (43)1-2b+2b{βz(f*g)(z)p(f*g)(z)(1-β)((f*g)(z)zp)α1-2b+2b+βz(f*g)(z)p(f*g)(z)((f*g)(z)zp)α}=1+2bp(αp+β)ap+1gp+1z+1bp(αp+2β)×{2ap+2gp+2+(α-1)ap+12gp+12}z2+2bp(αp+3β)×{ap+3gp+3+(α-1)ap+1ap+2gp+1gp+2+(α-1)(α-2)6ap+13gp+13}z3+. Also, we obtain (44)ϕ(w(z))=1+B1w1z+(B1w2+B1w12)z2+(B1w3+2B2w1w2+B3w13)z3+. Comparing the coefficients of z, z2, z3 and after simple calculations, we obtain (45)ap+1=bpB1w12gp+1(αp+β),ap+2=bpB12gp+2(αp+2β)×{w2-(bpB12(α-1)(αp+2β)2(αp+β)2-B2B1)w12},ap+3=bpB12gp+3(αp+2β){w3+q1w1w2+q2w13}. The remaining proof of the theorem is similar to the proof of Theorem 6.

Corollary 12.

For b=2, the results from (36) to (40) coincide with the results proved by Ramachandran et al.  for the class Rp,1,α,β,g(ϕ).

Corollary 13.

For b=2, α=0, and β=1,  the results from (36) to (40) coincide with the results obtained by Ali et al.  for the class Sp,g*(ϕ).

Corollary 14.

For b=2, p=1, α=0, and β=1, (46)g2=Γ(3)Γ(2-λ)Γ(3-λ)=22-λ,g3=Γ(4)Γ(2-λ)Γ(4-λ)=6(2-λ)(3-λ),B1=8π2,      B2=163π2, the results from (36) to (38) coincide with the results obtained by Srivastava and Mishra .

Acknowledgment

The work here is fully supported by LRGS/TD/2011/UKM/ICT/03/02.

Ali R. M. Ravichandran V. Seenivasagan N. Coefficient bounds for p-valent functions Applied Mathematics and Computation 2007 187 1 35 46 10.1016/j.amc.2006.08.100 MR2323552 Ma W. C. Minda D. A unified treatment of some special classes of univalent functions Proceedings of the Conference on Complex Analysis 1994 Cambridge, Mass, USA International Press 157 169 MR1343506 Owa S. Properties of certain integral operators Southeast Asian Bulletin of Mathematics 2000 24 3 411 419 10.1007/s100120050009 MR1811401 ZBL0980.30011 Ravichandran V. Gangadharan A. Darus M. Fekete-Szegő inequality for certain class of Bazilevic functions Far East Journal of Mathematical Sciences 2004 15 2 171 180 MR2133419 Chen M. P. On the regular functions satisfying Re{f(z)/z}>ρ Bulletin of the Institute of Mathematics. Academia Sinica 1975 3 1 65 70 MR0399470 ZBL0313.30012 Keogh F. R. Merkes E. P. A coefficient inequality for certain classes of analytic functions Proceedings of the American Mathematical Society 1969 20 8 12 MR0232926 10.1090/S0002-9939-1969-0232926-9 ZBL0165.09102 Prokhorov D. V. Szynal J. Inverse coefficients for (α,β)-convex functions Annales Universitatis Mariae Curie-Skłodowska A 1981 35 125 143 MR763927 Ramachandran C. Sivasubramanian S. Silverman H. Certain coefficient bounds for p-valent functions International Journal of Mathematics and Mathematical Sciences 2007 2007 11 46576 10.1155/2007/46576 MR2286001 Srivastava H. M. Mishra A. K. Applications of fractional calculus to parabolic starlike and uniformly convex functions Computers & Mathematics with Applications 2000 39 3-4 57 69 10.1016/S0898-1221(99)00333-8 MR1740907 ZBL0948.30018