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

Have there been papers dealing with random walks in $\mathbb{R}^n$ that are not on the lattice $\mathbb{Z}^n$? Instead of walking in one of the directions possible in $\mathbb{Z}^n$ with probability $1/2^n$, one would walk in the direction corresponding to points on $S^{n-1}$ with uniform probability. In the case of the lattice, the walk returns to the origin with probability $1$ for $n=1,2$ only. Do you know what happens in the continous case? Thanks.

share|cite|improve this question
In the real case you need to change the definition of recurrence: there's 0 probability of ever hitting any specific point, so you look instead at recurrence to a neighbourhood (i.e. a random walk is recurrent if for each initial point $x$ and for each $\epsilon > 0$, the walk almost surely returns to an $\epsilon$-neighbourhood of $x$. With this definition, the results are the same as in the discrete case: the random walk is recurrent in 1 and 2 dimensions; and transient in 3 dimensions or higher. – Anthony Quas Feb 17 '12 at 6:44
That was very helpful. Thank you very much. – alexei Feb 17 '12 at 7:55
A related result is Rayleigh's theorem: start at the origin in the plane and make $n$ steps of length one, with directions chosen independently and uniformly at each step. Then the probability of ending up at distance less than one from the origin is $1/(n+1)$. For an elegant recent proof by Bernardi, see – Richard Stanley Feb 17 '12 at 14:18
Look for some sources ... – Gerald Edgar Feb 17 '12 at 14:39
Thank you all very much for your helpful comments. – alexei Feb 18 '12 at 0:27

Your Answer


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

Browse other questions tagged or ask your own question.