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

Given two normal matrices $A,B\in M_n({\mathbb C})$ whose respective spectra are $(\alpha_{1},\ldots,\alpha_{n})$ and $(\beta_{1},\ldots,\beta_{n})$, is it true that $\det(A+B)$ belongs to the convex hull of the set of numbers $$\prod_{i=1}^n(\alpha_i+\beta_{\sigma(i)}),$$ as $\sigma$ runs over the set ${\mathfrak S}_n$ of permutations of $\{1,\ldots,n\}$ ?

Nota. It is known (see Exercise 101) that the trace of $AB$ belongs to the convex hull of the set of numbers $$\sum_{j=1}^n\alpha_{j}\beta_{\sigma(j)},\qquad\sigma\in {\mathfrak S}_n.$$

share|cite|improve this question
Interesting question, I hope some one will provide a solution. – Betrand Jun 29 '12 at 11:45

This claim is nothing but the well-known Marcus and de Oliveria conjecture, which has been open since 1973 or earlier.

Reference: Open Problems in Matrix Theory, X. Zhan.

For the simpler case of Hermitian matrices, the claim holds; a slightly more general case seems to be the paper "The validity of the Marcus-de Oliveira conjecture for essentially Hermitian matrices."

PS: you might want to add the "open problem" tag to your question.

share|cite|improve this answer
Thank you. I new that de Oliveira was concerned, but I did not succeed to find a paper of him about this, and I did not know about Markus' role. I'll wait to see if someone has some newer information. At least, I don't have the answer. As a rule, I don't ask questions of which I know the answer. – Denis Serre Oct 24 '10 at 17:35

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.