1
$\begingroup$

I would be interested in references on the topic of distances between probability measures that are singular with one another and not reduced to trivial ones. For example from here we know that total variation distance is not suitable for such a problem. In the end I am looking for a structure that would be fine enough to be able to address mutually singular Wiener measures in a non trivial way (for example it is well known that the Wiener measures associate with the two process driven by a BM are mutually singulars : $X^1_t =\sigma_1.W_t$ and $X^2_t =\sigma_2.W_t$, with $X^i_0=0, i=1,2$ as soon as $\sigma_1\not=\sigma_2$ so finding a meaningful distance between those measures is a matter of interest for me).

Best regards

Note : This a re-post from MathstackExchange where no answers nor useful comments where posted, the subject might be advanced enough to be worthy of Mathoverflow. If you feel that it is not the case not a problem as long as you can post useful elements at the MSE post.

Edit : I received an interesting idea from a contributor (Ilya who is also active on this forum) in the MSE post that I (tried to) developed there and which is elaborating on one of the metric (Wasserstein-Kantorovitch) pointed out in the article by Aryeh Kontorovich.

Improve this question
$\endgroup$
2
  • 5
    $\begingroup$ Here's a decent start: math.hmc.edu/~su/papers.dir/metrics.pdf $\endgroup$
    – Aryeh Kontorovich
    Commented Sep 13, 2016 at 10:19
  • 2
    $\begingroup$ In particular, any of those metrics which induce the weak topology on the space of probability measures should give you what you desire: that the law of $X_t$ varies continuously with $\sigma$. $\endgroup$
    – Nate Eldredge
    Commented Oct 3, 2016 at 17:56

0

Reset to default

You must log in to answer this question.