Skip to main content
15 events
when toggle format what by license comment
Apr 13, 2017 at 12:58 history edited CommunityBot
replaced http://mathoverflow.net/ with https://mathoverflow.net/
Sep 16, 2013 at 14:30 answer added Wolfgang timeline score: 2
Aug 11, 2013 at 11:50 answer added S. Carnahan timeline score: 6
Aug 9, 2013 at 23:41 answer added Igor Rivin timeline score: 1
Jul 10, 2013 at 19:40 history edited Wolfgang CC BY-SA 3.0
added 6x6 matrix with c(M)=9
Jul 5, 2013 at 7:34 comment added Wolfgang for $c(I_n)$, each cycle of length $d$ in a permutation contributes [(d-1)/2] conjugate pairs. $c(I_5)=118$ is correct.
Jul 5, 2013 at 7:33 comment added Wolfgang You are right for the first 3x3 one. I don't know what happened. Maybe a simple typo. Did you check my other ones?
Jul 4, 2013 at 22:38 answer added Felix Goldberg timeline score: 0
Jul 4, 2013 at 21:09 comment added Felix Goldberg I also get $c(I_{5})=8$ so this is all very odd - perhaps I am missing something?
Jul 4, 2013 at 20:50 comment added Felix Goldberg This is an interesting question but I have some trouble reproducing the examples. For instance, if I take the $3 \times 3$ matrix in the first example I get $c(M)=4$ instead of zero. (If you change places between the 1st and the 3rd rows, you get a matrix whose char. poly. is $x^{3}-x^{2}+2x-2$ which has two non-real roots. There is also another pair of non-real eigenvalues). Perhaps there is a bug in the OP's program.
Jul 4, 2013 at 19:43 history edited Wolfgang CC BY-SA 3.0
added Update 2
Jun 4, 2013 at 20:19 comment added Suvrit Wow, impressive!
Jun 4, 2013 at 20:18 history edited Wolfgang CC BY-SA 3.0
added a nice 4x4 one and found a 5x5 one with c(M)=0!
May 23, 2013 at 21:18 history edited Wolfgang CC BY-SA 3.0
added remark for n=4
May 23, 2013 at 20:58 history asked Wolfgang CC BY-SA 3.0