English

Diffeological statistical models, the Fisher metric and probabilistic mappings

Statistics Theory 2020-04-20 v2 Probability Statistics Theory

Abstract

In this note we introduce the notion of a CkC^k-diffeological statistical model, which allows us to apply the theory of diffeological spaces to (possibly singular) statistical models. In particular, we introduce a class of almost 2-integrable CkC^k-diffeological statistical models that encompasses all known statistical models for which the Fisher metric is defined. This class contains a statistical model which does not appear in the Ay-Jost-L\^e-Schwachh\"ofer theory of parametrized measure models. Then we show that for any positive integer kk the class of almost 2-integrable CkC^k-diffeological statistical models is preserved under probabilistic mappings. Furthermore, the monotonicity theorem for the Fisher metric also holds for this class. As a consequence, the Fisher metric on an almost 2-integrable CkC^k-diffeological statistical model PP(X)P \subset {\cal P}({\cal X}) is preserved under any probabilistic mapping T:XYT: {\cal X}\leadsto {\cal Y} that is sufficient w.r.t. PP. Finally we extend the Cram\'er-Rao inequality to the class of 2-integrable CkC^k-diffeological statistical models.

Keywords

Cite

@article{arxiv.1912.02090,
  title  = {Diffeological statistical models, the Fisher metric and probabilistic mappings},
  author = {Hông Vân Lê},
  journal= {arXiv preprint arXiv:1912.02090},
  year   = {2020}
}

Comments

16 p., final version, accepted to Journal Mathematics/MDPI

R2 v1 2026-06-23T12:35:51.350Z