16.3 SERIES SOLUTIONS ABOUT A REGULAR SINGULAR POINT
then at least one ofp(z)andq(z) is not analytic atz= 0, and in general we
should not expect to find a power series solution of the form (16.9). We must
therefore extend the method to include a more general form for the solution. In
fact, it may be shown (Fuch’s theorem) that there existsat least onesolution to
the above equation, of the form
y=zσ∑∞n=0anzn, (16.12)where the exponentσis a number that may be real or complex and wherea 0 =0
(since, if it were otherwise,σcould be redefined asσ+1 orσ+2 or···so as to
makea 0 = 0). Such a series is called a generalised power series orFrobenius series.
As in the case of a simple power series solution, the radius of convergence of the
Frobenius series is, in general, equal to the distance to the nearest singularity of
the ODE.
Sincez= 0 is a regular singularity of the ODE, it follows thatzp(z)andz^2 q(z)are analytic atz= 0, so that we may write
zp(z)≡s(z)=∑∞n=0snzn,z^2 q(z)≡t(z)=∑∞n=0tnzn,where we have defined the analytic functionss(z)andt(z) for later convenience.
The original ODE therefore becomes
y′′+s(z)
zy′+t(z)
z^2y=0.Let us substitute the Frobenius series (16.12) into this equation. The derivativesof (16.12) with respect tozare given by
y′=∑∞n=0(n+σ)anzn+σ−^1 , (16.13)y′′=∑∞n=0(n+σ)(n+σ−1)anzn+σ−^2 , (16.14)and we obtain
∑∞n=0(n+σ)(n+σ−1)anzn+σ−^2 +s(z)∑∞n=0(n+σ)anzn+σ−^2 +t(z)∑∞n=0anzn+σ−^2 =0.Dividing this equation through byzσ−^2 , we find
∑∞n=0[(n+σ)(n+σ−1) +s(z)(n+σ)+t(z)]anzn=0. (16.15)