@maxbiostat, your paper is behind a paywall! so I can't say anything about
it.

The strategy seems sound. When you can, send me a reprint so that I can
come up with more specific comments.

On Tue, Sep 1, 2015 at 1:48 PM, Luiz Max F. Carvalho <
[email protected]> wrote:

@fccoelho https://github.com/fccoelho , @lsbastos
https://github.com/lsbastos and @DVMath https://github.com/DVMath :

I propose to use the model in Carvalho, Bastos & Struchiner (2015)
http://download.springer.com/static/pdf/361/chp%253A10.1007%252F978-3-319-12454-4_18.pdf?originUrl=http%3A%2F%2Flink.springer.com%2Fchapter%2F10.1007%2F978-3-319-12454-4_18&token2=exp=1441126335%7Eacl=%2Fstatic%2Fpdf%2F361%2Fchp%25253A10.1007%25252F978-3-319-12454-4_18.pdf%3ForiginUrl%3Dhttp%253A%252F%252Flink.springer.com%252Fchapter%252F10.1007%252F978-3-319-12454-4_18*%7Ehmac=e342b03f4d5796c4ad352cf932a5a2c5d7cd5e8ed026f097ca6bfe79948f3027

It has three levels:

• 0. The hyperparameters of the thermodynamic functions;
• 1. The thermodynamic functions r(T) and K(T);
• 2. The population size P(t, T)

Problems:
Figure out a sensible propagation strategy: it probably doesn't make any
sense to propagate the uncertainty on the hyperparameters caused by
uncertainty on P(t,T), although it is certainly desirable to study how the
uncertainty about the hyperparameters influences the uncertainty on P(t,T).

An initial idea, thus, is to study only the prior on P(t,T) obtained by
combining:

Induced prior on P(t,T) from the hyperparameters + induced prior on P(t,T)
from the priors on the functions r(T) and K(T) and the actual prior on
P(t,T).
This would be nice because we would be combining uncertainty from three
levels and also because we could in principle apply our automatic methods
to choose the weights (alpha).

As a bonus, we could look into the combined prior for r(T) and K(T)
combining only:
the induced priors from the hyperparameters and the actual priors on r(T)
and K(T).

—
Reply to this email directly or view it on GitHub
#14.

Flávio Codeço Coelho

+55(21) 3799-5551
Professor
Escola de Matemática Aplicada
Fundação Getulio Vargas
Praia de Botafogo, 190 sala 312
Rio de Janeiro - RJ
22250-900
Brasil

from opinion_pooling.

@fccoelho check it now.

from opinion_pooling.

Closing for the time being. Model is really very complex and would be a nice test case. But we have enough on our plates as is.

from opinion_pooling.