Mathematics · Glossary

What is Radius of convergence?

Definition 11.2 University Mathematics — Year 2 · Chapter 11 — Power Series

The radius of convergence of anzn\sum a_n z^n is

R=sup{r0:(anrn) bounded}[0,+].R = \sup\{r \geq 0 : (a_n r^n) \text{ bounded}\} \in \intcc{0}{+\infty} .

By Lemma 11.1: absolute convergence for z<R\abs z < R (normal on compact subdisks), divergence — indeed unbounded terms — for z>R\abs z > R. On the boundary circle, anything can happen (Exercise 11.2). In practice RR is computed by d’Alembert’s ratio test on anzn\abs{a_n}\abs z^n or by comparison.

Examples

Example 11.3 (A radius with no ratio test)

What is the radius of sin(n)zn\sum \sin(n)\,z^n? The ratio sin(n+1)/sinn\abs{\sin(n+1)/\sin n} has no limit, but the definition works directly. R1R \geq 1: sinn1\abs{\sin n} \leq 1, so (sinnrn)(\sin n\cdot r^n) is bounded for every r<1r < 1 — indeed for r=1r = 1. R1R \leq 1: it suffices that sinn↛0\sin n \not\to 0. Suppose sinn0\sin n \to 0; the addition formula

sin(n+1)=sinncos1+cosnsin1\sin(n+1) = \sin n\cos 1 + \cos n\sin 1

would force cosn0\cos n \to 0 (solve for cosn\cos n: sin10\sin 1 \neq 0), contradicting sin2n+cos2n=1\sin^2 n + \cos^2 n = 1. So the terms sin(n)1n\sin(n)\,1^n do not tend to 00: the series diverges at z=1z = 1, and R1R \leq 1. Conclusion: R=1R = 1. Closing insight: the radius is a statement about anrn\abs{a_n}r^n being bounded — no limit of ratios is ever required, and boundedness arguments settle cases the ratio test cannot touch (compare the oscillating coefficients of Exercise 11.1).

Example 11.6 (Multiplying by 11x\frac{1}{1-x} sums the coefficients)

A Cauchy product against the geometric series has a memorable meaning: for any anxn\sum a_nx^n of radius R>0R > 0 and x<min(R,1)\abs x < \min(R, 1),

11xn0anxn=n0(k=0nak)xn:\frac{1}{1-x}\sum_{n\geq0}a_nx^n = \sum_{n\geq0}\Bigl(\sum_{k=0}^{n}a_k\Bigr)x^n :

multiplication by 11x\frac{1}{1-x} replaces coefficients by their partial sums (convolve with the all-ones sequence). Instance: ex1x=nsnxn\dfrac{\eu^x}{1-x} = \sum_n s_n x^n with sn=kn1k!s_n = \sum_{k\leq n}\frac{1}{k!}, the partial sums of e\eu — compare Exercise 11.11, where the same product with ex\eu^{-x} encodes the derangement counts. Closing insight: operations on power series are operations on coefficient sequences in disguise (multiply by 11x\frac1{1-x}: sum; multiply by xx: shift; differentiate: multiply by nn and shift) — a dictionary the generating-functions chapter will read fluently.

Example 11.20 (Binet’s formula, carried out)

Let φ=1+52\varphi = \frac{1+\sqrt5}{2} and ψ=152\psi = \frac{1-\sqrt5}{2}, the roots of X2=X+1X^2 = X + 1; since φ+ψ=1\varphi + \psi = 1 and φψ=1\varphi\psi = -1,

1xx2=(1φx)(1ψx).1 - x - x^2 = (1 - \varphi x)(1 - \psi x) .

Partial fractions: seeking x(1φx)(1ψx)=A1φx+B1ψx\frac{x}{(1-\varphi x)(1-\psi x)} = \frac{A}{1 - \varphi x} + \frac{B}{1 - \psi x}, the constant term gives A+B=0A + B = 0 and the xx-coefficient AψBφ=1-A\psi - B\varphi = 1, so A(φψ)=1A(\varphi - \psi) = 1: A=15=BA = \frac{1}{\sqrt5} = -B. Two geometric series later,

F(x)=15n0(φnψn)xnFn=φnψn5F(x) = \frac{1}{\sqrt5}\sum_{n\geq0} \bigl(\varphi^n - \psi^n\bigr)x^n \quad\Longrightarrow\quad F_n = \frac{\varphi^n - \psi^n}{\sqrt5}

by uniqueness of coefficients (Theorem 11.7). Since ψ<1\abs\psi < 1, the term ψn5\frac{\psi^n}{\sqrt5} has absolute value <12< \frac12: FnF_n is the nearest integer to φn5\frac{\varphi^n}{\sqrt5}. Closing insight: the radius 1φ\frac1\varphi of FF is the reciprocal of the dominant root — growth of coefficients and radius of convergence are the same information read in opposite directions.

Read in context →