Abstract
In this paper, our interest is devoted to study the convex combinations of the form \((1-\lambda)f+\lambda g\) where \(\lambda \in(0,1)\) of biholomorphic mappings on the Euclidean unit ball \({\mathbb B}^n\) the case of several complex variables. Starting from a result proved by S. Trimble [26] and then extended by P.N. Chichra and R. Singh [3, Th. 2] which says that if f is starlike such that \(Re[f'(z)]>0\) then \((1-\lambda)z+\lambda f(z)\) is also starlike, we are interested to extend this result to higher dimensions.
In the first part of the paper, we construct starlike convex combinations using the identity mapping on \({\mathbb B}^n\) and some particular starlike mappings on \({\mathbb B}^n\).
In the second part of the paper, we define the class \(L_{\lambda}^∗B^n\) and prove results involving convex combinations of normalized locally biholomorphic mappings and Loewner chains. Finally, we propose a conjecture that generalize the result proved by Chichra and Singh.
Authors
Eduard Stefan Grigoriciuc
Babes-Bolyai” University, Faculty of Mathematics and Computer Science, Cluj-Napoca, Romania
”Tiberiu Popoviciu” Institute of Numerical Analysis, Romanian Academy, Cluj-Napoca, Romania
Keywords
Biholomorphic mappings; Convex sums; Starlike mappings; Herglotz vector field; Loewner chains
Paper coordinates
E.S. Grigoriciuc, On some convex combinations of biholomorphic mappings in several complex variables, Filomat 36 (16) (2022), 5503-5519;
https://doi.org/10.2298/FIL226503G
About this paper
Journal
Filomat
Publisher Name
Faculty of Sciences and Mathematics, University of Nis, Serbia
Print ISSN
2406-0933
Online ISSN
google scholar link
[1] S.D. Bernardi, Convex and Starlike Univalent Functions, Trans. Amer. Math. Soc. 135 (1969), 429–446.
[2] D.M. Campbell, A survey of properties of the convex combination of univalent functions, Rocky Mountain J. Math. 5 (1975), 475–492.
[3] P. Chichra, R. Singh, Convex sum of univalent functions, J. Austral. Math. Soc. 14 (1972), 503–507.
[4] P.L. Duren, Univalent Functions, Springer-Verlag, New York, 1983.
[5] M. Elin, S. Reich, D. Shoikhet, Complex Dynamical Systems and the Geometry of Domains in Banach Spaces, Dissertationes Math. (Rozprawy Mat.) 427 (2004), 62 pp.
[6] I. Graham, H. Hamada, G. Kohr, Parametric Representation of Univalent Mappings in Several Complex Variables, Canad. J. Math. 54 (2002), 324–351.
[7] I. Graham, H. Hamada, G. Kohr, Radius problems for holomorphic mappings on the unit ball in Cn, Math. Nachr. 279 (2006), 1474–1490.
[8] I. Graham, H. Hamada, G. Kohr, M. Kohr, Extreme points, support points and the Loewner variation in several complex variables, Sci. China Math. 55(7) (2012), 1353–1366.
[9] I. Graham, G. Kohr, Geometric Function Theory in One and Higher Dimensions, Marcel Dekker Inc., New York, 2003.
[10] I. Graham, G. Kohr, M. Kohr, Loewner Chains and the Roper-Suffridge Extension Operator, J. Math. Anal. Appl. 247 (2000), 448–465.
[11] I. Graham, G. Kohr, M. Kohr, Loewner chains and parametric representation in several complex variables, J. Math. Anal. Appl. 281 (2003), 425–438.
[12] D.J. Hallenbeck, T.H. MacGregor, Linear Problems and Convexity Techniques In Geometric Function Theory, Pitman, Boston, 1984.
[13] H. Hamada, G. Kohr, Loewner chains and quasiconformal extension of holomorphic mappings, Ann. Polon. Math. 81 (2003), 85–100.
[14] H. Hamada, G. Kohr, Quasiconformal extension of biholomorphic mappings in several complex variables, J. Anal. Math. 96 (2005), 269–282.
[15] W.K. Hayman, Research Problems in Function Theory, The Athlone Press, London, 1967.
[16] G. Kohr, Basic Topics in Holomorphic Funcions of Several Complex Variables, Cluj University Press, Cluj-Napoca, 2003.
[17] T. Matsuno, On starlike theorems and convexlike theorems in the complex vector space, Sci. Rep. Tokyo Kyoiku Daigaku Sect. A 5 (1955), 88–95.
[18] T.H. MacGregor, The univalence of a linear combination of convex mappings, J. London Math. Soc. 44 (1969), 210–212.
[19] E.P. Merkes, On the convex sum of certain univalent functions and the identity function, Rev. Colombiana Math. 21 (1987), 5–12.
[20] J.A. Pfaltzgraff, Subordination Chains and Univalence of Holomorphic Mappings in Cn, Math. Ann. 210 (1974), 55–68.
[21] J.A. Pfaltzgraff, T.J. Suffridge, Close-to-starlike holomorphic functions of several complex variables, Pacif. J. Math. 57 (1975), 271–279.
[22] C. Pommerenke, Univalent Functions, Vandenhoeck and Ruprecht, Gottingen, 1975.
[23] S. Reich, D. Shoikhet, Nonlinear Semigroups, Fixed Points and the Geometry of Domains in Banach Spaces, World Scientific Publisher, Imperial College Press, London, 2005.
[24] K. Roper, T.J. Suffridge, Convexity properties of holomorphic mappings in Cn, Trans. Amer. Math. Soc., 351 (1999), 1803–1833.
[25] T.J. Suffridge, Starlikeness, convexity and other geometric properties of holomorphic maps in higher dimensions, Lecture Notes in Math., 599 (1976), 146–159.
[26] S. Trimble, The convex sum of convex functions, Math. Z. 109 (1969), 112–114.
On Some Convex Combinations of Biholomorphic Mappings in Several Complex Variables
Abstract
In this paper, our interest is devoted to study the convex combinations of the form , where , of biholomorphic mappings on the Euclidean unit ball in the case of several complex variables. Starting from a result proved by S. Trimble trimble and then extended by P.N. Chichra and R. Singh (chichra-singh, , Theorem 2) which says that if is starlike such that , then is also starlike, we are interested to extend this result to higher dimensions. In the first part of the paper, we construct starlike convex combinations using the identity mapping on and some particular starlike mappings on . In the second part of the paper, we define the class and prove results involving convex combinations of normalized locally biholomorphic mappings and Loewner chains. Finally, we propose a conjecture that generalize the result proved by Chichra and Singh.
1 Introduction
Let denote the space of complex variables with the Euclidean inner product and the Euclidean norm , for all . Also, let denote the Euclidean unit ball in . In the case of one complex variable, the unit disc is denoted by .
Let denote the set of all holomorphic mappings from into . If , we say that is normalized if and , where is the complex Jacobian matrix of at and is the identity operator in . Let
be the set of all normalized biholomorphic mappings on .
A mapping is called convex (starlike) if its image is a convex (respectively, starlike with respect to the origin) domain in . We denote by
the class of normalized convex mappings on and by
the class of normalized starlike mappings on . In the case of one complex variable, the sets , and are denoted by , and .
If , we say that is locally biholomorphic on if , for all , where , for all . We denote by
the set of all normalized locally biholomorphic mappings on . If , then is denoted by .
Another important class of normalized holomorphic functions on the unit disc is the Carathéodory class (for details, one may consult (duren, , Chapter 2), (graham-kohr1, , p. 27) or (pommerenke, , Chapter 2)), denoted by
In the case of several complex variables, we use the family
of normalized holomorphic mappings on the Euclidean unit ball. It is important to mention that the class plays the role of the Carathéodory family in . This class will be very important in the section that contains remarks about Loewner chains and Herglotz vector fields. For more details, one may consult graham-hamada-kohr1 , graham-hamada-kohr-kohr1 , graham-kohr1 , kohr1 , pfaltzgraff1 and reich-shoikhet .
Next, we recall the notions of Loewner chain and Herglotz vector field on the Euclidean unit ball in . We will use these notions to prove that (under some particular assumptions) the convex combination of two Loewner chains is also a Loewner chain.
Definition 1.1
(see e.g. (graham-kohr1, , Definition 8.1.2) or (graham-kohr-kohr2, , Definition 1.1)): A mapping is said to be a Loewner chain (normalized univalent subordination chain) if the following conditions hold:
-
1.
, for all ;
-
2.
, for all .
Definition 1.2
(see e.g. (graham-kohr1, , Chapter 8)): Let be a mapping. We say that is a Herglotz vector field if the following conditions hold:
-
1.
, for all ;
-
2.
is measurable on , for all .
The following theorem gives a sufficient condition for a mapping to be a Loewner chain (see (graham-hamada-kohr1, , Lemma 1.6), graham-kohr-kohr2 , (hamada-kohr2, , Lemma 2.3) or (pfaltzgraff1, , Theorem 2.2)).
Theorem 1.3
Let be a mapping which satisfies the following conditions:
-
1.
, and , for all ;
-
2.
is locally Lipschitz continuous on the interval locally uniformly with respect to .
Assume that there exists a Herglotz vector field such that
Moreover, assume that is a normal family on . Then is a Loewner chain.
The connection between the class and the Loewner chains in the case of one complex variable is given by a result due to Pommerenke (see e.g. (graham-kohr1, , Theorem 3.1.8)) which says that any function can be embedded as the first element of a Loewner chain (i.e. for each , there exists a Loewner chain such that , for all ).
This result is no longer true for the class , and by this reason I. Graham, H. Hamada and G. Kohr (see graham-hamada-kohr1 ) defined the class
of normalized univalent mappings which have parametric representation on . It is clear that (see pommerenke ), but , for (see graham-hamada-kohr1 and graham-kohr1 ). For details, one may consult also graham-hamada-kohr1 , (graham-kohr1, , Chapter 8) and graham-kohr-kohr2 .
2 Remarks on convex combinations
An interesting fact about the class of normalized univalent functions on the unit disc in is that the class is not convex. Namely, starting from two normalized univalent functions on the unit disc even the average of these functions does not necessarily belong to . To show this, we present two examples in the case of one complex variable (see e.g. duren , hallenbeck-macgregor or macgregor ) and one example on the Euclidean unit ball in (see e.g. graham-kohr1 or kohr1 ).
Example 2.1
In Example 2.1, the functions and are not only normalized and univalent - they are even starlike on the unit disc . However, the function is not starlike on (in fact, is not even univalent on ). Hence, , but because . Another important example was given by MacGregor in (macgregor, , Section 3). He proved that the linear combination of two convex functions is not necessarily univalent on the unit disc.
Example 2.2
In the previous example, are convex functions, but is not univalent on because there exists a point with , where , such that (for the complete proof, one may consult (macgregor, , Section 3)).
Next, we can extend the statement of Example 2.1 to the case of several complex variables. For , we obtain the following example (considered in (graham-kohr1, , Problem 6.2.3) and (kohr1, , Problem 4.3.4)):
Example 2.3
(see (graham-kohr1, , Problem 6.2.3)): Let
Then is not starlike on the Euclidean unit ball . In fact does not belong to .
Remark 2.4
We can also analyze the convex combination between the mapping , for and one of the mappings or , for . Similar arguments like in the previous examples will show us if the mapping is starlike or not on the Euclidean unit ball .
Although linear combinations of univalent functions are not always univalent (for more details about these results, one may consult campbell , hallenbeck-macgregor , hayman or macgregor ), there exist subclasses of the class that satisfy this condition (see chichra-singh , merkes or trimble in the case ). The goal of this paper is to extend in the case of several complex variables a result proved by P.N. Chichra and R. Singh in (chichra-singh, , Theorem 2) for the case of one complex variable. This result shows that the convex combination between a starlike function with positive real part of the derivative and the identity function is also starlike on the unit disc , as it follows:
Theorem 2.5
Let . If and Re, for all , then
(1) |
is starlike with respect to zero in and Re, for all .
In the case of several complex variables we start with some particular forms of the mapping in order to construct convex combinations which are starlike on the Euclidean unit ball . Like in the case of result proved by Chichra and Singh, we also consider convex combinations between a starlike mapping and the identity map in . First, we prove a general result for normalized locally bihomomorphic mappings on which satisfies some additional conditions in order to obtain the starlikeness of the convex combination. Then, in the final part, we propose a conjecture which generalize Theorem 2.5 in the case of several complex variables.
3 Preliminary results
Next we present some important results the will be used in the proofs of the main results from this paper. We recall, without proofs, the analytical characterization of starlikeness in (proved by Matsuno, see matsuno or (graham-kohr1, , Theorem 6.2.2); see also elin-reich-shoikhet and suffridge ), the connection between Loewner chains and starlike mappings (see (graham-kohr1, , Corollary 8.2.3) or (pfaltzgraff-suffridge, , Corollary 2)) and also some important criteria for univalence in .
Theorem 3.1
Let be a locally biholomorphic mapping such that . Then is starlike if and only if
(2) |
Using Theorem 3.1 we can prove that a locally biholomorphic mapping with is starlike on the Euclidean unit ball . Another important characterization of starlikeness was given by Pfaltzgraff and Suffridge (see (graham-kohr1, , Corollary 8.2.3) or (pfaltzgraff-suffridge, , Corollary 2)) in terms of Loewner chains.
Theorem 3.2
Let be a normalized locally biholomorphic mapping on . Then is starlike on if and only if is a Loewner chain.
We end this section with two important results that ensure the univalence of a normalized holmorphic mapping on the Euclidean unit ball . The first result, proved by Suffridge in (suffridge, , Theorem 7), is a version of the Noshiro-Warschawski’s univalence criteria (see e.g. (duren, , Theorem 2.16) for the case ) in the case of several complex variables.
Theorem 3.3
(see (suffridge, , Theorem 7)): Let be a normalized holomorphic mapping such that
(3) |
Then is univalent on .
Another important criteria for univalence in is presented in the following result proved by I. Graham, H. Hamada and G. Kohr in (graham-hamada-kohr2, , Lemma 2.2) (see also hamada-kohr1 and hamada-kohr2 ).
Theorem 3.4
(see (graham-hamada-kohr2, , Lemma 2.2)): Let be a normalized holomorphic mapping such that
(4) |
Then . In particular, is univalent on .
Taking into account that , for (see graham-hamada-kohr1 ), it is clear that there is an important difference between Theorem 3.3 (which assures us the univalence of a mapping on ) and Theorem 3.4 (which assures us that a mapping admits parametric representation on ), i.e. there exist normalized holomorphic mappings on that satisfy condition (3), but do not satisfy condition (4).
Such an example, that shows us that is a proper subclass of for , is presented in (graham-kohr1, , Example 8.3.21).
4 Univalence of convex combinations in
In view of the results presented in the previous section we can prove some criteria for univalence of a convex combination of normalized holomorphic mappings on the Euclidean unit ball . In fact, we can obtain a condition for a convex combination to be a mapping which has parametric representation on .
Lemma 4.1
Let be a normalized holomorphic mapping such that
for all , with . Also let
Then is univalent on .
Proof 4.2.
Clearly, is a normalized holomorphic mapping. Moreover,
and then
for all , with . According to Theorem 3.3 we obtain that is univalent on . Since is also normalized, it means that .
Remark 4.3.
Notice that, in view of the previous proof, we obtain that if Re, for all and with , then Re, for all and with .
Let . We say that a mapping is a homogenous polynomial of degree if there exist an --linear operator such that . For details, one may consult graham-kohr1 or kohr1 . Recall that we define the norm of the operator by .
Lemma 4.4.
Let be a normalized holomorphic mapping such that
for all and
Then . In particular, is univalent on .
Proof 4.5.
Frist, it is clear that is a normalized holomorphic mapping on . Moreover,
and then
for all . In view of Theorem 3.4 (see (graham-hamada-kohr2, , Lemma 2.2)) we obtain that , so is also univalent on .
Lemma 4.6.
Let be a normalized holomorphic mapping such that , for all . Also let
If , then .
Proof 4.7.
Since , for all , we deduce that
where we used the fact that is a homogenous polynomial of degree . On the other hand,
Hence,
Then we obtain
Consequently,
and in view of Theorem 3.4 we deduce that .
5 Particular starlike mappings on the Euclidean unit ball in
In this section we present the first important result of this paper together with some examples to illustrate how this result can be applied in several particular cases. We begin this section with a well-known result related to starlike mappings on the Euclidean unit ball (see graham-kohr1 ).
Lemma 5.1.
Let be of the form , for all .
-
1.
If , then .
-
2.
If, in addition, Re, for all , then Re, for all and with .
Proof 5.2.
Indeed, if , then (see e.g. (graham-kohr1, , Problem 6.2.5) or (kohr1, , Example 4.3.4)) and this completes the first part of the proof.
For the second part of the lemma, we have
and
for all and with and this completes the proof.
According to Lemma 5.1 we can obtain a first version of Theorem 2.5 in the case of several complex variables. However, in this case we have a particular form of the mapping (it has on each component a starlike function of one variable). The following result is very simple and its proof is immediate.
Proposition 5.3.
Let and let be such that Re, for all . Also, let , for all . Then
(5) |
is starlike, for all and . Moreover, Re, for all and with . In particular, is univalent on .
Proof 5.4.
It is clear that the mapping used in the previous result has a very particular form (has on each component a starlike function of one complex variable). However, we can obtain similar results for an arbitrary starlike mapping on the Euclidean unit ball.
In the following examples (considered also in elin-reich-shoikhet , (graham-kohr-kohr1, , Example 3.5), (hamada-kohr1, , Example 3.4), roper-suffridge or (suffridge, , Examples 3 and 7)) we use arbitrary starlike mappings to construct starlike mappings (as convex combinations on ). On the other hand, we can obtain starlike mappings of the form (5) on using starlike mappings that do not satisfy condition Re, for all and with .
Example 5.5.
Let and be given by
(6) |
with . According to (roper-suffridge, , Example 5), we know that . Moreover,
so
Then if and only if . In particular, this is true for and . Hence, .
Example 5.6.
Let and be given by
(7) |
with . According to (roper-suffridge, , Example 6), we know that . Moreover,
so
Then if and only if . In particular, this is true for and . Hence, .
It is important to mention here that the results contained in Examples 5.5 and 5.6 can be directly verified. However, we can obtain starlike mappings as convex combinations of two starlike mappings such that at least one of them does not satisfy the condition Re, for all and with .
Remark 5.7.
In the following example (considered also in graham-kohr1 and roper-suffridge ) we use a convex mapping on to construct a starlike univalent mapping on the Euclidean unit ball . The condition also ensures the univalence of the mapping on .
Example 5.8.
Let and be given by
(8) |
with . Then is starlike on , for all and .
Proof 5.9.
According to (roper-suffridge, , Example 7) we know that . Moreover,
Then because . In addition, and
for all , with and . On the other hand,
(9) |
and
Then
and
(10) |
for all , with and . Hence, is a starlike univalent mapping on the Euclidean unit ball .
6 A general result on convex combinations of locally biholomorphic mappings in several complex variables
Next, we present another suggestive result which can be seen as a second version of Theorem 2.5 in the case of several complex variables. We can consider the following result as a generalization of the theorem proved by Chichra and Singh (see (chichra-singh, , Theorem 2)).
Theorem 6.1.
Let and . Also let be a normalized locally biholomorphic mapping such that
(11) |
and
(12) |
for all . Consider be given by
(13) |
Then , for all .
Proof 6.2.
In order to prove that , for all , it is enough to show that is locally biholomorphic on , and
Since is normalized on it follows that
Moreover,
for all , in view of relation (11). According to Theorem 3.4 the above inequality assures us that is univalent on . Obviously, in particular, is locally biholomorphic on the Euclidean unit ball . Moreover, the relation
implies that is invertible, i.e. there exists the inverse operator . Finally, the inequality
is equivalent to
Hence, in view of relation (12), we obtain that
for all . According to Theorem 3.1 we conclude that , for all .
Remark 6.3.
It is clear that
-
•
if , then ;
-
•
if , then ,
for all and then .
7 The class
Taking into account the main result from the previous section, we can define a class of normalized locally biholomorphic mappings on the Euclidean unit ball that satisfies conditions from Theorem 6.1. Hence, the convex combination between the identity map and a function from this class will be a starlike mapping on the Euclidean unit ball .
Definition 7.1.
Let us consider and . We say that if is normalized locally biholomorphic on and satisfies
-
;
-
Re, for all .
In view of the above definition, we denote the class
Remark 7.2.
Note that as the identity mapping given by , for all belongs to .
Next, we present an example (considered also in elin-reich-shoikhet , (graham-kohr-kohr1, , Example 3.5) or (suffridge, , Examples 3 and 7)) of a mapping that has been used in Example 5.8 and which satisfies also the conditions and from Definition 7.1. This means that and then we can construct a starlike mapping which is a convex combination between the identity mapping and the mapping .
Example 7.3.
Let and be given by
(14) |
where . Then . In particular, , where , for all and .
Proof 7.4.
Let and . Then is normalized locally biholomorphic on and
If we denote
then
Let us consider such that . It follows that
and if and only if . Since and , we obtain that condition from Definition 7.1 is satisfied. On the other hand,
and
(15) |
Let us denote
Then
for all . Hence, condition from Definition 7.1 is also satisfied and then we conclude that . In particular, in view of Theorem 6.1 we obtain that , where is given by relation (13).
Remark 7.5.
In the second part of this section, let us to refer to the case . Consider , and a normalized locally univalent function on the unit disc in . Then
-
1.
Condition can be written in one of the following form
(16) where is the disc with center and radius . The smallest disc can be constructed for (in this case, we obtain the disc of center and radius ). On the other hand, for , it is clear that Re for implies condition (16), but the converse result is not necessarily true.
-
2.
Condition can be written in one of the following form
or
(17) for all with . Clearly, if and Re for , then the above condition is satisfied (for details, one may consult bernardi or chichra-singh ). But again, the converse result, is not necessarily true.
Hence, we conclude that if is sufficiently small and is a function with the property that Re for , then and take place, but the converse implication is not necessarily true.
-
3.
In view of previous remarks we can define the class
for the case of one complex variable, where and .
Remaining in the case we obtain the following result:
Proposition 7.6.
Let and . Consider the function be given by
Then .
Proof 7.7.
Since we deduce that , and (in fact, ). Moreover,
for all , where . Hence, in view of the analytical characterization of starlikeness in (see (graham-kohr1, , Theorem 2.2.2)) we obtain that .
Question 7.8.
What is the connection between the starlikeness of the mapping and conditions and for ? For sure, there will be no equivalence between the conditions, but the question would be whether the implication from the case is true.
8 Remarks on Loewner chains
Another interesting approach to the class is that in terms of Loewner chains. We can prove that starting from a function we can easily construct an associated Loewner chain according to Theorem 6.1 and the characterization of starlikeness with Loewner chains given by Theorem 3.2. In particular, we can obtain a Loewner chain that is the convex combination of another two Loewner chains.
Proposition 8.1.
Let . If , then
(18) |
is a Loewner chain, for all and .
Proof 8.2.
Remark 8.3.
In the following remark we replace the mapping with a starlike mapping on the Euclidean unit ball. However in order to obtain a Loewner chain (which is also a convex combination of the identity mapping and a starlike mapping on ) we still need the assumption from Definition 7.1. According to this remark we deduce that in our context, for this condition is very important.
Remark 8.4.
Let and be such that , for all . Also consider the mapping be given by
(19) |
In this case, is the convex combination of two Loewner chains
and
Moreover, is holomorphic on , and , for all . On the other hand, is locally Lipschitz continuous on locally uniformly with respect to .
According to relation (19) we have that
for all and . Let us denote given by
for all , and . Since , we deduce that is a normal family on (see (duren, , Chapter 1) and kohr1 ).
In order to prove that is a Loewner chain (according to Theorem 1.3) we have to construct a Herglotz vector field such that
(20) |
Using the assumption , for all , we deduce that
for all . Then the operator is invertible on and we can consider the inverse operator on the Euclidean unit ball . In view of this remark and relation (20) we obtain
or equivalently,
and hence
(21) |
for all and . Clearly, is measurable on , for all because is constant with respect to and then it remains to prove that .
For simplicity let us consider . Since is invertible, it follows that is holomorphic on , and , for all . Indeed,
for all . On the other hand, if we denote , then
where , for all . Taking into account the previous relations we obtain the mapping
for all and . Now it is clear that , for all . Moreover, after some computations we deduce that
Proposition 8.5.
Let and be such that
(23) |
and
(24) |
for all . Then given by (19) is a Loewner chain. In particular, the Loewner chain is the convex combination of two Loewner chains.
Remark 8.6.
Let , and be such that , for all . In view of relation (17) we obtain that condition (24) is
(25) |
If we denote , for all , then is holomorphic and normalized on . In fact, in view of the assumption , for all , we obtain that . Next, let us define given by
According to previous remarks, we deduce that , for all and is measurable on the interval , for all , where
is the Carathéodory class in the case of one complex variable (for details, one may consult (duren, , Chapter 2), (graham-kohr1, , p. 27) or (pommerenke, , Chapter 2)).
If we consider given by , then , and , for all . Moreover,
Hence, taking into account the version of Theorem 1.3 (see (graham-kohr1, , Theorem 3.1.13)), we conclude that is a Loewner chain in .
9 Conjecture related to Chichra-Singh’s result in several complex variables
In this last section we propose a conjecture (for the case of several complex variables) which generalize Theorem 2.5 proved by Chichra and Singh in chichra-singh (in the case of one complex variable).
Conjecture 9.1.
Let . If and Re, for all and with , then
is a starlike mapping on . Moreover, Re, for all and with . In particular, is univalent on .
Remark 9.2.
In the case of one complex variable, the statement of Conjecture 9.1 is true, as it reduces to Theorem 2.5 obtained by Chichra and Singh in chichra-singh .
Acknowledgments
The author thanks to his first doctoral supervisor, Professor PhD Gabriela Kohr, for her advice and very useful suggestions during the preparation of this paper. The author is also grateful to his current PhD advisor, Professor PhD Mirela Kohr, for her continuous support and encouragement. The author thanks the referee(s) for carefully reading the manuscript and providing helpful suggestions.
References
- (1) S.D. Bernardi, Convex and Starlike Univalent Functions, Trans. Amer. Math. Soc. 135 (1969), 429–446.
- (2) D.M. Campbell, A survey of properties of the convex combination of univalent functions, Rocky Mountain J. Math. 5 (1975), 475–492.
- (3) P. Chichra, R. Singh, Convex sum of univalent functions, J. Austral. Math. Soc. 14 (1972), 503–507.
- (4) P.L. Duren, Univalent Functions, Springer-Verlag, New York, 1983.
- (5) M. Elin, S. Reich, D. Shoikhet, Complex Dynamical Systems and the Geometry of Domains in Banach Spaces, Dissertationes Math. (Rozprawy Mat.) 427 (2004), 62 pp.
- (6) I. Graham, H. Hamada, G. Kohr, Parametric Representation of Univalent Mappings in Several Complex Variables, Canad. J. Math. 54 (2002), 324–351.
- (7) I. Graham, H. Hamada, G. Kohr, Radius problems for holomorphic mappings on the unit ball in , Math. Nachr. 279 (2006), 1474–1490.
- (8) I. Graham, H. Hamada, G. Kohr, M. Kohr, Extreme points, support points and the Loewner variation in several complex variables, Sci. China Math. 55(7) (2012), 1353–1366.
- (9) I. Graham, G. Kohr, Geometric Function Theory in One and Higher Dimensions, Marcel Dekker Inc., New York, 2003.
- (10) I. Graham, G. Kohr, M. Kohr, Loewner Chains and the Roper-Suffridge Extension Operator, J. Math. Anal. Appl. 247 (2000), 448–465.
- (11) I. Graham, G. Kohr, M. Kohr, Loewner chains and parametric representation in several complex variables, J. Math. Anal. Appl. 281 (2003), 425–438.
- (12) D.J. Hallenbeck, T.H. MacGregor, Linear Problems and Convexity Techniques In Geometric Function Theory, Pitman, Boston, 1984.
- (13) H. Hamada, G. Kohr, Loewner chains and quasiconformal extension of holomorphic mappings, Ann. Polon. Math. 81 (2003), 85–100.
- (14) H. Hamada, G. Kohr, Quasiconformal extension of biholomorphic mappings in several complex variables, J. Anal. Math. 96 (2005), 269–282.
- (15) W.K. Hayman, Research Problems in Function Theory, The Athlone Press, London, 1967.
- (16) G. Kohr, Basic Topics in Holomorphic Funcions of Several Complex Variables, Cluj University Press, Cluj-Napoca, 2003.
- (17) T. Matsuno, On starlike theorems and convexlike theorems in the complex vector space, Sci. Rep. Tokyo Kyoiku Daigaku Sect. A 5 (1955), 88–95.
- (18) T.H. MacGregor, The univalence of a linear combination of convex mappings, J. London Math. Soc. 44 (1969), 210–212.
- (19) E.P. Merkes, On the convex sum of certain univalent functions and the identity function, Rev. Colombiana Math. 21 (1987), 5–12.
- (20) J.A. Pfaltzgraff, Subordination Chains and Univalence of Holomorphic Mappings in , Math. Ann. 210 (1974), 55–68.
- (21) J.A. Pfaltzgraff, T.J. Suffridge, Close-to-starlike holomorphic functions of several complex variables, Pacif. J. Math. 57 (1975), 271–279.
- (22) C. Pommerenke, Univalent Functions, Vandenhoeck and Ruprecht, Gottingen, 1975.
- (23) S. Reich, D. Shoikhet, Nonlinear Semigroups, Fixed Points and the Geometry of Domains in Banach Spaces, World Scientific Publisher, Imperial College Press, London, 2005.
- (24) K. Roper, T.J. Suffridge, Convexity properties of holomorphic mappings in , Trans. Amer. Math. Soc., 351 (1999), 1803–1833.
- (25) T.J. Suffridge, Starlikeness, convexity and other geometric properties of holomorphic maps in higher dimensions, Lecture Notes in Math., 599 (1976), 146–159.
- (26) S. Trimble, The convex sum of convex functions, Math. Z. 109 (1969), 112–114.