A. inference in thermal systems

Some rules for inference about thermal systems

by P. Fraundorf of Physics & Astronomy/Center for NanoScience at University of Missouri -St Louis (63121)

We show here (based on class notes provided by Edwin T. Jaynes) how rules for ``taking the best guess", in both parameter-estimation and in model-selection, can be put into a form that is familiar to thermodynamics experts (who've pioneered ``analog" application of these tools) well before any observations of physical systems themselves are taken into account. This might provide cross-disciplinary context for folks learning about thermal physics for the first time. It remains to be seen if this particular exercise is of value to researchers applying Bayesian inference in other areas (cf. Burnham2002) as well. 

Sections of this paper available so far include: