Timeline for why do we need to specify to symmetric matrix when defining real positive definite matrix? [closed]
Current License: CC BY-SA 3.0
6 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Dec 4, 2012 at 18:22 | history | closed |
Ryan Budney Pietro Majer Emil Jeřábek Chris Godsil Andreas Blass |
not a real question | |
Dec 4, 2012 at 18:07 | comment | added | Tobias Fritz | Can you provide the link to the Wikipedia page? If it really says that "This property implies that $M$ is an Hermitian matrix", then this is simply wrong and should be corrected. For example, if $M$ is anti-hermitian, then $z^T Mz=0$ for all $z$. | |
Dec 4, 2012 at 17:49 | comment | added | Pietro Majer | The partial order for symmetric matrices is defined as $A\ge B$ iff $A-B$ is positive. If you try to define it for all matrices, most properties would be destroyed. Check the proofs. | |
Dec 4, 2012 at 17:47 | answer | added | arbitUser1401 | timeline score: 1 | |
Dec 4, 2012 at 17:43 | comment | added | kjetil b halvorsen | The point is that $z^T M z$ defines a quadratic form. If $M$ is not symmetric it can be replaced with $1/2(M+M^T)$ and with that matrix in place of $M$, exactly the same quadratic form results! check it. | |
Dec 4, 2012 at 17:08 | history | asked | user29665 | CC BY-SA 3.0 |