Misc. Stats Topics Discussion > Partial derivative w.r.t. standard deviation
First of all, allow me to apologize for the extremely delayed nature of my response. Unfortunately, my forum notifications were being sent to an inactive email, so I didn't become aware of new posts for a long time. With that said, I'm afraid your question is a little unclear to me. Can you provide a little more context?
February 23, 2012 |
Jeremy Taylor
Jeremy Taylor
In some papers (two standard deviations), they use partial derivative with respect to one standard deviation assuming that the correlation is constant. However, I could not find any mathematical condition (or reasons) that the correlation coefficient constant when one standard deviation increases.
Can someone tell me what are mathematical conditions for it?
For example,
V=(s1)^2+(s2)^2+2*(s1)(s2)(col),
(dV/dS1)=2(s1)+2(s2)(col)----X
Under what conditions, above X is true?
Thanks