Remarks on a theorem of Perron

From MaRDI portal
(Redirected from Publication:890201)



Abstract: For a linear differential equation with a mild condition on its singularities, we discuss generalized continued fractions converging to expressions in its solutions and their derivatives. In the case of an order two linear differential equation, this is the logarithmic derivative of the holomorphic solution near a singularity.


Let be given a linear differential equation of order two \[ y={q_0}(z)y+{q_1}(z)y,\;{q_0},{q_1}\in \mathbb{C}(z).\tag{1} \] The \(n\)-th derivative of \(y\) can be presented as \[ {y^{(n)}}={q_{0,n}}(z)y+{q_{1,n}}(z)y,{q_{i,n}}\in \mathbb{C}(z),\;i=0,1. \] Let \(S=\left\{ {t_1},{t_2},\dots,{t_r} \right\}\) be the set of singular points of (1) in \({\mathbb C}\). With \(S\) one can associate the open sets \({U_1},{U_2},\dots ,{U_r}\) defined by \[ {U_i}:=\left\{ z\in \mathbb{C} \mid |z-{t_i}| < |z-{t_j}|,\;j=1,2,\dots,r,\;j\neq i \right\}. \] The authors say that (1) is generic if \(\forall {t_i}\in S\) there is a basis \({y_1},{y_2}\) of the space of solutions of (1) such that near \({t_i}\) the solution \({y_2}\) extends holomorphically to \({t_i}\) and the solution \({y_1}\) does not extend. For the equation (1) the following theorem is proved in the paper: Theorem. For a generic Fuchsian differential equation (1), the fraction \(\frac{q_{0,n}(z)}{q_{1,n}(z)}\) converges uniformly in compact subsets of \(U_i\) to \(-\frac{y'_2}{y_2}\) , where \(y_2\) is the unique (up to multiplication by a constant) holomorphic solution of (1) in \(U_i\). Then, the authors show that \(\frac{q_{0,n}(z)}{q_{1,n}(z)}\) can be written as the continued fraction. So they get a generalization of a theorem of Perron for the hypergeometric equation (see [\textit{O. Perron}, Die Lehre von den Kettenbrüchen. Band II. Analytisch-funktionentheoretische Kettenbrüche. 3. verbesserte und erweiterte Auflage. Stuttgart: B. G. Teubner Verlagsgesellschaft (1957; Zbl 0077.06602)]).











This page was built for publication: Remarks on a theorem of Perron

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q890201)