| |
BGonline.org Forums
A better approch?
Posted By: Timothy Chow In Response To: A better approch? (Maik Stiebler)
Date: Thursday, 5 November 2009, at 6:52 a.m.
You're right that I haven't fully justified my suspicion that the computational effort would be less, and variance reduction more easily handled, using the approach I'm suggesting. At the moment it is just an intuition.
As for the posterior distribution being normal, I still don't understand what you're saying. The slides you linked to assume a Gaussian prior and not a uniform prior, and the latter is what you said in your previous post. Without any calculation, it should be clear that if you start with a uniform distribution and make an observation, the distribution can't suddenly change to a perfectly Gaussian distribution.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.