1
$\begingroup$

What do we call a set of Gaussian distributions with the same covariance but different means?

Is there a particular term for that?

I mean the Gaussians are like:

$N(\mu_1,\Sigma), N(\mu_2,\Sigma), N(\mu_3,\Sigma)...$

$\endgroup$
4
  • $\begingroup$ Do you mean with the same variances and with all covariances between pairs of different variables the same? $\endgroup$
    – Glen_b
    Mar 17, 2016 at 2:32
  • $\begingroup$ @Glen_b yes, all the same, the only difference is the mean. $\endgroup$
    – dontloo
    Mar 17, 2016 at 2:33
  • $\begingroup$ Wait, I thought you were talking about common values within a single $\Sigma$. So the common $\Sigma$ is an arbitrary covariance matrix? $\endgroup$
    – Glen_b
    Mar 17, 2016 at 2:40
  • $\begingroup$ @Glen_b sorry I misunderstood, yes $\Sigma$ is just an arbitrary covariance matrix. $\endgroup$
    – dontloo
    Mar 17, 2016 at 2:42

1 Answer 1

2
$\begingroup$

When different multivariate distributions have the same covariance matrix, you could describe them as having "common covariance structure" or "a common covariance matrix, $\Sigma$".

You could perhaps try a term like "equi-covariant" but I don't think I've seen such a term used, and it might not be understood.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.