User talk:Wjastle

Polygamma in Beta Distribution Higher Moments section
Wjastle,

Could you please double check your math?

Instead of your posted result:


 * $$\operatorname{E}(\log^2 X) = \psi^2(\alpha) - \psi^2(\alpha + \beta)+[\psi^{(1)}(\alpha)-\psi^{(1)}(\alpha+\beta)]^2,$$

I get:


 * $$\operatorname{E}(\log^2 X) = (\psi(\alpha) - \psi(\alpha + \beta))^2+\psi^{(1)}(\alpha)-\psi^{(1)}(\alpha+\beta),$$

same problem for the other result

Thanks,

Dr. J. Rodal (talk) 01:04, 11 August 2012 (UTC)

Yes you are right, I got the notation confused: mixed powers and orders of poly-gammas.


 * I just added "Should the Mean and the Variance based on Ln[X] (rather than X) be included in the statistical property box?" to the Beta Distribution talk page. It will be interesting to hear your views on this subject, and for our Wikipedia community to arrive at a consensus. Thanks Dr. J. Rodal (talk) 15:30, 11 August 2012 (UTC)