Dear colleagues, Encouraged by your recent exchanges, which show that the topic of Hans' New Year Lecture is far from exhausted, I would like to think a bit more on the fundamental change from Frequentist to Bayesian statistics. Hans writes:
“On the one hand each individual agent assembles the totality of her experiences (experimenting, reading, talking, calculating...) into a web of probability assignments that is as coherent and comprehensive as possible. That's the easy part, and, as usual, physicists have picked it as their domain. But the hard part is the effort of agents to correlate their private experiences -- i.e. to communicate with each other in order to develop a common scientific worldview. Agent A's description of an experience serves as input for updating B's personal probability assignments via Bayes' law. And this is done through language as well as math.” (Hans mail from Saturday, January 18, 2014 6:47 PM) Reading the above I conclude that QBist change of perspective is not only relevant for quantum physics, or physics in general. It is relevant for all sciences based on observations and experiments. And indeed, among others, brain researchers are using Bayesian statistics. However, there are brain researchers arguing for the necessity of going beyond Bayes: http://www.mdpi.com/2078-2489/3/2/175 Beyond Bayes: On the Need for a Unified and Jaynesian Definition of Probability and Information within Neuroscience by Christopher D. Fiorillo Are there any comments to this claim? Would Jaynesian statistics make a difference for Qbism? I would like to learn more. With best wishes, Gordana http://www.mrtc.mdh.se/~gdc/<http://www.mrtc.mdh.se/%7Egdc/>
_______________________________________________ fis mailing list fis@listas.unizar.es https://webmail.unizar.es/cgi-bin/mailman/listinfo/fis