Premium
On Hacking's fiducial theory of inference
Author(s) -
Sharma Shyam S.
Publication year - 1980
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.2307/3315234
Subject(s) - notation , hacker , terminology , inference , mathematical notation , computer science , fiducial marker , fiducial inference , basis (linear algebra) , mathematical theory , calculus (dental) , theoretical computer science , artificial intelligence , mathematics , bayesian inference , frequentist inference , bayesian probability , linguistics , arithmetic , computer security , philosophy , medicine , geometry , dentistry , physics , quantum mechanics
Abstract Fisher's theory of fiducial inference is known to lead to mathematical inconsistencies. Hacking (1965) puts forward a new basis for fiducial inference. In this paper we have presented Hacking's theory retaining his original terminology but replacing his notation by more usual mathematical notation. His condition of irrelevance is derived in an explicit mathematical form, making it more comprehensible to statisticians. Further an example is constructed to show that Hacking's theory also leads to mathematical inconsistencies.