Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Earlier today on MathWorld (see eq. 17) I ran across the following expression, which gives a generating function for the divisor function $\sigma_k(n)$: $$\sum_{n=1}^{\infty} \sigma_k (n) x^n = \sum_{n=1}^{\infty} \frac{n^k x^n}{1-x^n}. \tag{1}$$

(The divisor function $\sigma_k(n)$ is defined by $\sigma_k(n) = \sum_{d|n} d^k$.)

How would one go about proving Equation (1)?

(For reference, I ran across this while thinking about this question asked earlier today. The sum in that question is a special case of the sum on the right side of Equation (1) above.)

share|cite|improve this question
up vote 7 down vote accepted

Switching the order of summation, we have that

$$\sum_{n=1}^{\infty}\sigma_{k}(n)x^{n}=\sum_{n=1}^{\infty}x^{n}\sum_{d|n}d^{k}=\sum_{d=1}^{\infty}d^{k}\sum_{n:\ d|n}^{\infty}x^{n}.$$ From here, applying the formula for the geometric series, we find that the above equals $$\sum_{d=1}^{\infty}d^{k}\sum_{n=1}^{\infty}x^{nd}=\sum_{d=1}^{\infty}d^{k}\frac{x^{d}}{1-x^{d}}.$$

Such a generating series is known as a Lambert Series. The same argument above proves that for a multiplicative function $f$ with $f=1*g$ where $*$ represents Dirichlet convolution, we have $$\sum_{n=1}^\infty f(n)x^n =\sum_{k=1}^\infty \frac{g(k)x^k}{1-x^k}.$$

share|cite|improve this answer
Well done. Thanks, Eric! – Mike Spivey Jan 9 '13 at 4:16
I don't think $f$ and $g$ even have to be multiplicative for that identity to hold. – Greg Martin Jan 9 '13 at 6:12

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.