lsbastos
commented
9 years ago Yeup,
The correct one:
################# Thus \begin{align} \pi(\mu)&= t(\boldsymbol\alpha)\prod_{i=0}^{K}f_i(\mu; m_i, v_i)^{\alphai}\ &\propto \prod{i=0}^{K} \left[ \exp \left(-\frac{(\mu-m_i)^2}{2v_i} \right) \right]^{\alphai} \ &\propto \exp\left[-\frac{1}{2}\left{\mu^2 \sum{i=0}^K\frac{\alpha_i}{vi} - 2 \mu \sum{i=0}^K \frac{\alpha_i m_i}{vi} \right} \right] % \ %&\propto \exp \left( -\frac{1}{2}\frac{(\mu-m^)^2}{v^} \right) \end{align} which yields a normal distribution with parameters and $m^* = \frac{\sum{i=0}^K w_i mi}{\sum{i=0}^K wi}$ and $v^* = [\sum{i=0}^K w_i]^{-1}$, where $w_i = \alpha_i/v_i$. ################
- Fiz as correções nas equacoes (29-31)
- Mudei a variancia s_i^2 para v_i, só para tirar o "^2".
- Vc me explicou, mas eu me esqueci como usar o github. Um dia eu aprendo. :-)
___@_'-' Leonardo S. Bastos, PhD Researcher at Scientific Computing Program (PROCC) Oswaldo Cruz Foundation - Fiocruz Homepage: http://www.procc.fiocruz.br/Members/lsbastos CV Lattes: http://lattes.cnpq.br/5241799121437269 Phone: +55 21 3836 1112 (office) Alternative emails: lsbastos.fiocruz@gmail.com; lsbastos@pq.cnpq.br
2015-08-26 14:04 GMT-03:00 Luiz Max F. Carvalho notifications@github.com:
there's a minus sign that shouldn't be there.
Re-do the calcs
— Reply to this email directly or view it on GitHub https://github.com/maxbiostat/opinion_pooling/issues/11.
there's a minus sign that shouldn't be there.
Re-do the calcs