The existence of an analytic solution of an infinite order differential equation and the nature of its domain of analyticity
Sbornik. Mathematics, Tome 9 (1969) no. 1, pp. 53-71
Voir la notice de l'article provenant de la source Math-Net.Ru
We consider the equation
\begin{equation}
Ly\equiv\sum_{k=0}^\infty a_k y^{(k)}(z)=f(z)
\end{equation}
under the assumption that the characteristic function $a(z)=\sum_{k=0}^\infty a_kz^k$ is an entire function which does not grow faster than an exponential function of minimal type (that is, $a(z)\in[1,0]$). If $G$ is an arbitrary domain, we let $E(G)$ denote the set of all functions which are analytic in $G$, and we let $L(E(G))$ be the image of $E(G)$ under the operator $Ly$ acting from $E(G)$ into $E(G)$. We let $W(y)$ denote the complete Weierstrass domain of existence of an arbitrary analytic function $y(z)$.
Theorem 1. If $G$ is a finite convex domain, then $L(E(G))= E(G)$.
\smallskip
Theorem 2. If $G$ is not a simply connected domain, then $L(E(G))$ is a proper subset of $E(G)$.
\smallskip
Theorem 3. Let the function $y(z)$ be analytic at $z_0\in W(f)$ and satisfy
equation $(1)$ in a neighborhood of this point. Then:
a) if $W(f)$ is simply connected, then $W(y)$ is simply connected;
b) if $W(f)$ is convex, then $W(y)$ is convex.
Assertion 3b) for the case where $f(z)$ is an entire function extends a theorem of Polya.
We note an important qualitative difference between linear equations of finite and infinite order. Namely, under the assumptions of Theorem 3 for a finite problem we know that $W(y)=W(f)$, but for an infinite problem we can always find a solution $y_1(z)$ for which $W(y_1)$ is a proper subset of $W(f)$.
The following theorem is specifically for equations of infinite order, and does not have a finite analog.
Theorem 4. {\it If $G$ is a domain which is not convex and $a(z)$ is a transcendental entire function in the class $[1,0],$ then there exists an operator $L_1y=\sum_{k=0}^\infty b_ky^{(k)}(z)$ with characteristic function $a_1=a(e^{i\varphi_2}z),$ $\varphi_2\in[0,2\pi],$ such that $L_1(E(G))$ is a proper subset of $E(G)$}.
We note here that if $a(z)$ is a polynomial and $G$ is a finite, simply connected domain, then $L(E(G))=E(G)$.
In this work we shall find necessary and sufficient conditions for solvability of equation (1) in $E(G)$ for a given right-hand side $f(z)\in E(G)$. We establish a connection between solvability conditions and certain interpolation problems for exponential functions. We shall examine certain examples.
Bibliography: 15 titles.
@article{SM_1969_9_1_a1,
author = {Yu. F. Korobeinik},
title = {The existence of an analytic solution of an infinite order differential equation and the nature of its domain of analyticity},
journal = {Sbornik. Mathematics},
pages = {53--71},
publisher = {mathdoc},
volume = {9},
number = {1},
year = {1969},
language = {en},
url = {http://geodesic.mathdoc.fr/item/SM_1969_9_1_a1/}
}
TY - JOUR AU - Yu. F. Korobeinik TI - The existence of an analytic solution of an infinite order differential equation and the nature of its domain of analyticity JO - Sbornik. Mathematics PY - 1969 SP - 53 EP - 71 VL - 9 IS - 1 PB - mathdoc UR - http://geodesic.mathdoc.fr/item/SM_1969_9_1_a1/ LA - en ID - SM_1969_9_1_a1 ER -
%0 Journal Article %A Yu. F. Korobeinik %T The existence of an analytic solution of an infinite order differential equation and the nature of its domain of analyticity %J Sbornik. Mathematics %D 1969 %P 53-71 %V 9 %N 1 %I mathdoc %U http://geodesic.mathdoc.fr/item/SM_1969_9_1_a1/ %G en %F SM_1969_9_1_a1
Yu. F. Korobeinik. The existence of an analytic solution of an infinite order differential equation and the nature of its domain of analyticity. Sbornik. Mathematics, Tome 9 (1969) no. 1, pp. 53-71. http://geodesic.mathdoc.fr/item/SM_1969_9_1_a1/