0
$\begingroup$

Is there any characterization of the set of entire functions $f(z) $ such that $\Re(f(z)) \geq \Re(\overline{f(\bar{z})})$ for all $z\in \mathbb{C}^{+} $?

($\Re$ stands for the real part)

Edit: I was thinking in writing $f(z) $ in the form

$$f(z) =f(x+iy) =u(x, y) +iv(x, y) $$

and

$$\overline{f(\bar{z})}=\overline{f(x-iy)} =u(x, - y) - iv(x, - y) $$

so, it is like finding all such continuous functions $u$ for which $$u(x, y) \geq u(x, - y), \;\;\forall\; y>0, x\in\mathbb {R} $$

$\endgroup$
0

1 Answer 1

2
$\begingroup$

Your last inequality implies that $v(x,y):=u(x,y)-u(x,-y)\geq 0,\; y>0$, and similarly $v(x,y)\leq 0,\; y<0$. Since $v$ is harmonic, this easily implies that $v(x,y)=cy$ for some real constant $c$. Now $w(x,y):=u(x,y)-cy/2$ will satisfy $w(x,y)=w(x,-y)$ This implies $(\partial w/\partial y)(0,y)=0$ and by Cauchy-Riemann, the conjugate function $w^*$ to $w$ is constant on the real line. Since $(u+iu^*)(z)=w+iw^*(z)+ciz/2$, we obtain a characterization: these are exactly those entire functions whose imaginary part is of the form $a+bx$ on the real line, where $a$ and $b$ are constants, $b$ is real. (And your original inequality is in fact always equality). Or in other words, the general form of such $f$ is $f=g+a+biz$ where $g$ is real on the real line.

Remark. I used the fact that a harmonic function which is positive for $y>0$ and negative for $y<0$ must be $cy$. This is easy to prove, but is also a special case of a general theorem describing meromorphic functions with positive real part in the upper half-plane and negative real part in the lower half-plane, see for example Levin, Distribution of zeros of entire functions, Ch. VI, sect. 1, Them 1.

$\endgroup$
6
  • $\begingroup$ thank you, this is true, and contains more. The function $f(z) =-iz=-ix+y$ satisfy the condition above but it is not real for real $z$. $\endgroup$
    – Guest
    Commented Dec 26, 2019 at 3:48
  • $\begingroup$ @Guest: thanks, I corrected. $\endgroup$ Commented Dec 26, 2019 at 4:20
  • $\begingroup$ the content $b$ depends on the choice of the function $g$, is this wright? $\endgroup$
    – Guest
    Commented Dec 26, 2019 at 10:31
  • $\begingroup$ @Guest: I made one more small correction, hopefully the last one. Now $b$ does not depend on he choice of $g$, only on $f$ itself. If you impose the condition that $a$ is pure imaginary, (include real part of $a$ into $g$) then the representation becomes canonical: $g$, $a$ and $b$ depend only of $f$. $\endgroup$ Commented Dec 26, 2019 at 13:09
  • $\begingroup$ Ok. I only have a question about the first line when you defined $v$: I know that $v(x, y)=\Im(f(z)) =\frac{f(z) - \overline {f(z)}} {2i}$. So, is your $v$ is the same as the imaginary part of $f$? According to what you wrote it is like having $v(x, y)=\frac{f(z) - \overline {f(\bar{z} )}} {2i}$!!! $\endgroup$
    – Guest
    Commented Dec 26, 2019 at 13:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .