In my studies of random matrices, I recently came across this challenge:

Let $X \in \mathbb{R}^{m \times n}$ and $Y \in \mathbb{R}^{n \times k} $ be two independent Gaussian random matrices, i.e., with entries independently sampled from $\mathcal{N}(0,1)$ (a normal distribution with zero mean and unit variance). We define the positive orthant probability $$p_+(m,n,k) \triangleq P(\forall i,j: (XY)_{ij} >0).$$ Question: How does $\log_2 p_+(m,n,k)$ behave asymptotically, to leading order, in the limit $ n \rightarrow \infty$ and $\alpha m = \beta k = n $, for some positive constants $\alpha$ and $\beta$?

Note that if $n\rightarrow \infty$ but $m$ and $k$ remain fixed, then from the central limit theorem $\frac{1}{\sqrt{n}}XY$ becomes a Gaussian matrix with independent entries, and then we get $\log_2 p_+(m,n,k) = -mk$. However, I wish to know if this asymptotic behavior persists when $m$ and $k$ are of similar magnitude as $n$. In other words, for what positive constants $a, b$ and $c$, do we have

$$\lim_{\underset{\alpha m = \beta k = n}{n\rightarrow\infty}} \frac{1}{m ^{a }k^{b}} \log_2 p_+(m,n,k) = -c \,.$$

Is it $a=b=c=1$, like in the case of fixed $m$ and $k$, or something else? I guess the answer should depend on $\alpha$ and $\beta$.

If no exact answer will be given to my question, I'll award the answer to whoever supplies the tightest upper bound (non-trivial, in which $a+b>1$).

Thanks in advance!

  • 1
    $\begingroup$ FYI: If you don't get responses here, you -may- want to try Math StackExchange. Interesting question, nevertheless. (+1) $\endgroup$ – Clarinetist Sep 22 '16 at 12:15
  • $\begingroup$ @Clarinetist : will do thanks, I hope someone helps though $\endgroup$ – kroner Sep 22 '16 at 12:16
  • 2
    $\begingroup$ This is exactly the same question as this one on MathOverflow: mathoverflow.net/questions/249969/… $\endgroup$ – David R Sep 22 '16 at 12:23
  • $\begingroup$ @DavidRobertson : yes I figured perhaps it would get more attention here $\endgroup$ – kroner Sep 22 '16 at 12:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.