scientific article; zbMATH DE number 3921704

From MaRDI portal
Publication:3696288





[For the entire collection see Zbl 0567.00012.] A historical review of \textit{L. Boltzmann}'s work on entropy and its relationship to the asymptotic theory of statistics, especially to prediction, is discussed. The concept of entropy provides a proof of the objectivity of the log likelihood as a measure of the goodness of fit of a statistical model. From the predictive point of view the magic number 2 can be explained by use of the AIC statistic (Akaike information criterion). In the author's words the predictive point of view and the concept of entropy can produce a unifying view of statistics, this is also practical and productive.... Bayesian modeling will often be an improvement over non- Bayesian approaches.




Cited in
(23)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3696288)