reflections on the probability space induced by moment conditions with implications for Bayesian Inference [slides]
Here are the slides of my incoming discussion of Ron Gallant’s paper, tomorrow.
Here are the slides of my incoming discussion of Ron Gallant’s paper, tomorrow.
December 4, 2014 at 5:57 pm
Another random thought: This has the structure of a moment problem, so surely the answer to the existence question basically comes down to a “matrix completion” problem: For any sample size, can the sequence of Hankel matrices can be completed to a positive definite infinite Hankel matrix?
The answer would probably be “sometimes and here are some uncheckable conditions”.
December 4, 2014 at 4:33 pm
Typo in the references on slide 17 (the ABC, EP/INLA etc… a year is mis-typed)
December 4, 2014 at 4:37 pm
How kind of you! Thanks!