12 Parabolic Cylinder FunctionsProperties

§12.2 Differential Equations

Contents
  1. §12.2(i) Introduction
  2. §12.2(ii) Values at z=0
  3. §12.2(iii) Wronskians
  4. §12.2(iv) Reflection Formulas
  5. §12.2(v) Connection Formulas
  6. §12.2(vi) Solution U¯⁡(a,x); Modulus and Phase Functions

§12.2(i) Introduction

PCFs are solutions of the differential equation

12.2.1 d2wdz2+(a⁢z2+b⁢z+c)⁢w=0,

with three distinct standard forms

12.2.2 d2wdz2−(14⁢z2+a)⁢w=0,
12.2.3 d2wdz2+(14⁢z2−a)⁢w=0,
12.2.4 d2wdz2+(ν+12−14⁢z2)⁢w=0.

Each of these equations is transformable into the others. Standard solutions are U⁡(a,±z), V⁡(a,±z), U¯⁡(a,±x) (not complex conjugate), U⁡(−a,±i⁢z) for (12.2.2); W⁡(a,±x) for (12.2.3); Dν⁡(±z) for (12.2.4), where

12.2.5 Dν⁡(z)=U⁡(−12−ν,z).

All solutions are entire functions of z and entire functions of a or ν.

For real values of z (=x), numerically satisfactory pairs of solutions (§2.7(iv)) of (12.2.2) are U⁡(a,x) and V⁡(a,x) when x is positive, or U⁡(a,−x) and V⁡(a,−x) when x is negative. For (12.2.3) W⁡(a,x) and W⁡(a,−x) comprise a numerically satisfactory pair, for all x∈ℝ. The solutions W⁡(a,±x) are treated in §12.14.

In ℂ, for j=0,1,2,3, U⁡((−1)j−1⁢a,(−i)j−1⁢z) and U⁡((−1)j⁢a,(−i)j⁢z) comprise a numerically satisfactory pair of solutions in the half-plane 14⁢(2⁢j−3)⁢π≤ph⁡z≤14⁢(2⁢j+1)⁢π.

§12.2(ii) Values at z=0

12.2.6 U⁡(a,0) =π212⁢a+14⁢Γ⁡(34+12⁢a),
12.2.7 U′⁡(a,0) =−π212⁢a−14⁢Γ⁡(14+12⁢a),
12.2.8 V⁡(a,0) =π⁢212⁢a+14(Γ⁡(34−12⁢a))2⁢Γ⁡(14+12⁢a),
12.2.9 V′⁡(a,0) =π⁢212⁢a+34(Γ⁡(14−12⁢a))2⁢Γ⁡(34+12⁢a).

§12.2(iii) Wronskians

§12.2(iv) Reflection Formulas

For n=0,1,…,

12.2.13 U⁡(−n−12,−z)=(−1)n⁢U⁡(−n−12,z),
12.2.14 V⁡(n+12,−z)=(−1)n⁢V⁡(n+12,z).

§12.2(v) Connection Formulas

12.2.17 2⁢π⁢U⁡(−a,±i⁢z)=Γ⁡(12+a)⁢(e∓i⁢π⁢(12⁢a−14)⁢U⁡(a,z)+e±i⁢π⁢(12⁢a−14)⁢U⁡(a,−z)).
12.2.18 2⁢π⁢U⁡(a,z)=Γ⁡(12−a)⁢(e∓i⁢π⁢(12⁢a+14)⁢U⁡(−a,±i⁢z)+e±i⁢π⁢(12⁢a+14)⁢U⁡(−a,∓i⁢z)),
12.2.19 U⁡(a,z)=±i⁢e±i⁢π⁢a⁢U⁡(a,−z)+2⁢πΓ⁡(12+a)⁢e±i⁢π⁢(12⁢a−14)⁢U⁡(−a,±i⁢z).

§12.2(vi) Solution U¯⁡(a,x); Modulus and Phase Functions

When z (=x) is real the solution U¯⁡(a,x) is defined by

12.2.21 U¯⁡(a,x)=Γ⁡(12−a)⁢V⁡(a,x),

unless a=12,32,…, in which case U¯⁡(a,x) is undefined. Its importance is that when a is negative and |a| is large, U⁡(a,x) and U¯⁡(a,x) asymptotically have the same envelope (modulus) and are 12⁢π out of phase in the oscillatory interval −2⁢−a<x<2⁢−a. Properties of U¯⁡(a,x) follow immediately from those of V⁡(a,x) via (12.2.21).

In the oscillatory interval we define

where F⁡(a,x) (>0), θ⁡(a,x), G⁡(a,x) (>0), and ψ⁡(a,x) are real. F or G is the modulus and θ or ψ is the corresponding phase.

For properties of the modulus and phase functions, including differential equations, see Miller (1955, pp. 72–73). For graphs of the modulus functions see §12.3(i).