Let $S_n$ be the space of real $n \times n$ symmetric matrices and let $S_n^+$ be the convex cone of positive semidefinite matrices in $S_n$. The extremal rays of this cone correspond to the positive semidefinite matrices of rank $1$.

My problem is that I have a subcone $C$ of $S_n^+$ which is given by (finitely many) inequalities, which are not all linear (quadratic for example). In $C$ I have another subcone $C'$ given by generators which correspond to rank $1$ matrices. Since I would like to compare $C$ and $C'$, it would be nice to have inequalities defining $C'$. So my question is if there are known methods to determine such inequalities.


Any closed convex set in a locally convex topological vector space over $\mathbb R$ is the intersection of some family of half-spaces. In particular, any convex cone in ${\mathbb R}^{n \times n}$ is defined by a collection of linear inequalities (not necessarily equations). Moreover, since the space is separable, a countable collection will do. I don't know what else you would expect to say about it.

| cite | improve this answer | |
  • $\begingroup$ Sorry, I guess I was not clear with my question. I hope it is clearer now. $\endgroup$ – Max Sep 8 '11 at 23:13

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.