Answers are grounded in this paper and cite their sources.
PARABOLIC CYLINDER.117 equation, when n is unrestricted, is taken by Whittaker to be ) c -*-i(W (-t)-^ dt, and other solutions are D a {-z), Dn -\{+iz).When n is a positive integer, thenThe moat remarkable discovery which has been made concerning the function D n (z) is contained in Adamoff's* approximation, namely that, 'when n is a large integer and x is positive,and | S«(aO| < 3*35 ..., when n > 2 ; and ic,,(0) = 0 or £&/» (where -1 < 9 < 1), according as n is odd or even.In my former paper, I shewed that Adamoffs investigation would be extended to B lh {z), when z is complex, provided that (z)\ is not too large : and this result made it easy to give a formal proof of a theorem enunciated by Hermite and Whittaker, concerning the expansibility of an arbitrary analytic function in the form 2 a n D n (,?) :-a result analogous to Neumann's expansions in series of Legendre polynomials and Bessel functions.In this paper, I investigate various types of asymptotic formulae and expansions of D n (z), hy the method of steepest descents ; the results contained in the paper are therefore to be associated with those previously obtained by Debyet and by myself I for Bessel functions.It should be stated here that the method of steepest descents^ is a mode of discussing the integral f e" Jif) <p (s) ds by choosing a contour which passes through a zero of /' (s), and which * Ann.del'Inst.Poly technique de St. Petersbourg, t. 5 (1906), pp.127-143.More recently, Perron, Archiv der Math.u.Phys., Ser. 3, Vol.22 (1914), pp.329-340, has given some approximate formulae for the confluent hypergeometric function of which D n {z) is a special form, but the limits of error in the formulae are not investigated with the preciseness of the •earlier work of Adamoff.t Math.
No takes yet. Share an insight, caveat, or question.
G. N. Watson (1910) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: