Hello, again, Paul

> In effect, when forecasting snow, one should ignore priors and listen to The 
> Weather Channel.

Pooling forecasts (like yours and the Weather Channel's) can be done
in a variety of ways, even within arguable Bayesian orthodoxy. If you
believe the Weather Channel is well calibrated, and have no such
confidence in your own forecast, then adopting their forecast sounds
like a plan.

In any case, the Bayes' ratio that reconciles _your_ prior probability
to _their_ posterior probability wouldn't seem to be helpful to you.

Paul
_______________________________________________
uai mailing list
[email protected]
https://secure.engr.oregonstate.edu/mailman/listinfo/uai

Reply via email to