TOPICS
Search

Frobenius Method


The Frobenius method constructs solutions of a linear ordinary differential equation near a regular singular point by allowing a power series to be multiplied by a possibly nonintegral power of the displacement from the expansion point.

At an ordinary point x_0, expand y in a Taylor series about x_0. Commonly, the expansion point can be taken as x_0=0, resulting in the Maclaurin series

 y=sum_(n=0)^inftya_nx^n.
(1)

Plug y back into the ODE and group the coefficients by power. Now, obtain a recurrence relation for the nth term, and write the series expansion in terms of the a_ns. Expansions for the first few derivatives are

y=sum_(n=0)^(infty)a_nx^n
(2)
y^'=sum_(n=1)^(infty)na_nx^(n-1)
(3)
=sum_(n=0)^(infty)(n+1)a_(n+1)x^n
y^('')=sum_(n=2)^(infty)n(n-1)a_nx^(n-2)
(4)
=sum_(n=0)^(infty)(n+2)(n+1)a_(n+2)x^n.

If x_0 is a regular singular point of the ordinary differential equation

 P(x)y^('')+Q(x)y^'+R(x)y=0,
(5)

translate the independent variable so that x_0=0 and seek a Frobenius solution of the form

 y=x^ksum_(n=0)^inftya_nx^n,
(6)

where a_0!=0. Its derivatives are

y=x^ksum_(n=0)^(infty)a_nx^n
(7)
=sum_(n=0)^(infty)a_nx^(n+k)
y^'=sum_(n=0)^(infty)a_n(n+k)x^(k+n-1)
(8)
y^('')=sum_(n=0)^(infty)a_n(n+k)(n+k-1)x^(k+n-2).
(9)

Substitute into the ordinary differential equation and group by power to obtain a recurrence relation for the coefficients a_n. Setting the coefficient of the lowest power to zero gives the indicial equation, which determines the allowed values of k.

As an example, consider the Bessel differential equation

 x^2(d^2y)/(dx^2)+x(dy)/(dx)+(x^2-m^2)y=0.
(10)

Plugging (6) into (10) yields

 sum_(n=0)^infty(k+n)(k+n-1)a_nx^(k+n)+sum_(n=0)^infty(k+n)a_nx^(k+n)
 +sum_(n=2)^inftya_(n-2)x^(k+n)-m^2sum_(n=0)^inftya_nx^(n+k)=0.
(11)

The indicial equation, obtained by setting n=0, is then

 a_0[k(k-1)+k-m^2]=a_0(k^2-m^2)=0.
(12)

Since a_0!=0, k^2-m^2=0, so k=+/-m. Take m>=0 and choose k=m. The next coefficient gives

 a_1(2m+1)=0
(13)

(so a_1=0) and

 [a_nn(2m+n)+a_(n-2)]x^(m+n)=0
(14)

for n=2, 3, ..., so

 a_n=-1/(n(2m+n))a_(n-2)
(15)

for n>1. With a_0=1/[2^mGamma(m+1)], where Gamma is the gamma function, substitution into (6) gives the Bessel function of the first kind

 J_m(x)=sum_(j=0)^infty((-1)^j)/(j!Gamma(j+m+1))(x/2)^(2j+m).
(16)

For noninteger m, J_m(x) and J_(-m)(x) are linearly independent solutions on x>0. For integer m>=0, however,

 J_(-m)(x)=(-1)^mJ_m(x),
(17)

and a second linearly independent solution is the Bessel function of the second kind Y_m(x).

Fuchs's theorem guarantees at least one Frobenius solution at a regular singular point. Such a solution need not be an ordinary power series or a Laurent series, since its leading exponent can be nonintegral. If the roots of the indicial equation differ by an integer, including zero, a second linearly independent solution may contain a logarithm.


See also

Bessel Differential Equation, Frobenius Solution, Fuchs's Theorem, Indicial Equation, Ordinary Differential Equation, Regular Singular Point

Explore with Wolfram|Alpha

References

Arfken, G. "Series Solutions--Frobenius' Method." §8.5 in Mathematical Methods for Physicists, 3rd ed. Orlando, FL: Academic Press, pp. 454-467, 1985.Frobenius. "Ueber die Integration der linearen Differentialgleichungen durch Reihen." J. reine angew. Math. 76, 214-235, 1873.Ince, E. L. Ch. 5 in Ordinary Differential Equations. New York: Dover, 1956.National Institute of Standards and Technology. "Connection Formulas." §10.4 in Digital Library of Mathematical Functions. https://dlmf.nist.gov/10.4.National Institute of Standards and Technology. "Regular Singularities: Fuchs-Frobenius Theory." §2.7(i) in Digital Library of Mathematical Functions. https://dlmf.nist.gov/2.7#i.

Referenced on Wolfram|Alpha

Frobenius Method

Cite this as:

Weisstein, Eric W. "Frobenius Method." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/FrobeniusMethod.html

Subject classifications