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

How to show that for arbitrary( complex) trigonometric polynomials $P$ and $Q$ holds

$$\frac{1}{2\pi} \int_{-\pi}^{\pi} P(t)Q(mt)dt = \frac{1}{2\pi} \int_{-\pi}^{\pi} P(t)dt \frac{1}{2\pi} \int_{-\pi}^{\pi} Q(t)dt$$

always when $m \in \mathbb{Z}$ is sufficiently large.

Hint: Use information that

$$\frac{1}{2\pi} \int_{-\pi}^{\pi} e^{int}dt = 0\,\,,\,\,when \,\,n\neq 0\,\,\,,\,\, and \,\,\,\frac{1}{2\pi} \int_{-\pi}^{\pi} e^{int}dt = 1\,\,\, when\,\,n=0$$ Just some hint. I know that complex trigonometric polynomials is

$$\sum_{n=-m}^{m} c_n e^{int}$$

but I also know that if $\,f(t)=t\,$ and $\,g(t)=t+1\,$ , then

$$\int_{-\pi}^{\pi} f(t)g(t)dt \neq \int_{-\pi}^{\pi} f(t)dt \cdot \int_{-\pi}^{\pi} g(t)dt$$

So I doubt that it is true for complex trigonometric polynomial.

share|cite|improve this question
$w$ occurs in the premise, but not in the claim. $m$ occurs in the claim, but not in the premise. Copied from illegible handwriting? – Hagen von Eitzen Oct 21 '12 at 13:01
What is that $\,w\,$ there? Something related to both functions' periodicy? – DonAntonio Oct 21 '12 at 13:03
Without taking that still weird $\,w\,$ into account, I think your example proves the claim's utterly false... – DonAntonio Oct 21 '12 at 13:08
Yes I changed $w$ into $m$ as it should be. – alvoutila Oct 21 '12 at 16:58
up vote 1 down vote accepted

The functions $t \mapsto e^{i k t}$ for $k \in \mathbb{Z}$ form an orthonormal set on the interval $[-\pi, \pi]$ with inner product

$$ \langle Q, P \rangle = \frac{1}{2 \pi} \int_{-\pi}^{\pi}Q(t)\overline{P(t)}dt. $$

If $m$ is sufficiently large in your expression, the only term from $Q$ that contributes to the stated inner product is therefore its constant term. This constant term is expressed by the latter integral of $Q$ (the inner product of $Q$ with the constant function $\mathbf{1}$ if you wish). The complex conjugate that appears in the inner product is irrelevant: simply consider $\langle Q, \overline{P} \rangle$.

share|cite|improve this answer
How would you define both $P(t)$ and $Q(t)$? $$P(t)= \sum_{n=-m}^{m} a_n e^{int}$$ and $$Q(t)=\sum_{n=-m}^{m} a_n e^{int}$$? – alvoutila Oct 21 '12 at 19:54
@alvoutila like that but not necessarily with identical coefficients of course. So linear combinations over $\mathbb{C}$ of the basis functions $e^{i k t}$. – WimC Oct 22 '12 at 5:18

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.