MathOverflow is a question and answer site for professional mathematicians. 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

Let $x_1,x_2,...,x_k$ be irrational number,is it always true that: $\liminf_{n\rightarrow\infty} \sum_{i=1}^k (nx_i)=0$ (where $(x)$ denotes the fractional part of $x$)

If not,what are the necessary and sufficient conditions that {$x_i$} must satisfy so that $\liminf_{n\rightarrow\infty} \sum_{i=1}^k (nx_i)=0$

share|cite|improve this question
Oh sorry,I have made a mistake in the question.$(x)$ should be equal to $|x-[x]|$,where $[x]$ is the nearest integer to $x$. – Ben May 21 '12 at 1:19
up vote 4 down vote accepted

Here is a counterexample: $x_1 = \sqrt2, x_2 = 1-\sqrt2$. For $n>0$, $(nx_1) + (n x_2) =1$ so the limit infimum is $1$, not $0$.

I think the other answers assumed that you meant $x-[x]$, where $[x]$ is the nearest integer to $x$, instead of $(x)$.

If there is no rational dependency, then the multiples are dense in the torus $(\mathbb{R}/\mathbb{Z})^k$, so the limit infimum is $0$. If there is some rational dependency, there will always be small multiples (close to the origin in $(\mathbb{R}/\mathbb{Z})^k$), but these might not be positive.

Consider the closure of multiples of $\langle x_1,...,x_k \rangle$ in the torus $(\mathbb{R}/\mathbb{Z})^k$. This is a closed subgroup, hence a finite union of parallel flat subtori. For $(x)$, the condition you want is that the tangent space to this subgroup at the origin intersects the positive orthant.

share|cite|improve this answer

If you take any vector in Euclidean space, and look at its integer multiples modulo the integer lattice, we know what happens qualitatively. They are dense in a certain subtorus of the obvious torus. That's a version of Kronecker's theorem. Your question is about being near the origin infinitely often. So I say it's true.

share|cite|improve this answer
That is not exactly correct, since this only shows you are near a lattice point, but not necessarily near the origin. – Igor Rivin May 19 '12 at 13:52
Yes, there's the good point I missed that where the numbers are rationally dependent there is the geometry explained in another answer. – Charles Matthews May 20 '12 at 7:53

Yes. This is the subject of "simultaneous diophantine approximation", googling which will produce many results, but you can take a look at Doug Hensley's notes:

share|cite|improve this answer

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.