Revisiting the saddle-point method of Perron (Q1719665)

Let $p$ and $q$ are holomorphic functions and integration is along a contour $\mathcal{C}$. The accurate estimation of \[ \int_{\mathcal{C}} e^{n\cdot p(z)}\cdot q(z) dz \] as $N \to \infty$ is the main problem which is considered in the article. If the contour can be moved to pass through a saddle-point of $p(z)$ so that $\mathbf{Re}\, p(z)$ achieves its maximum on $\mathcal{C}$ there, then the complete asymptotic expansion of integral may be given quite explicitly. This was established by \textit{O. Perron} in his paper [Münch. Ber. 191--220 (1917; JFM 46.0538.03)]. Two proofs of the key result are given in the present paper. The first proof is a reworking of Perron's original proof. The second proof based on Laplace's method extends the approach of \textit{F. W. J. Olver} [Asymptotics and special functions. New York - London: Academic Press (1974; Zbl 0303.41035)]. New results include more precise error terms and bounds for the expansion coefficients. Perron's original examples in greater details are treated and a new application to the asymptotics of Sylvester waves are given.