Physics > Data Analysis, Statistics and Probability
[Submitted on 28 Mar 2002 (v1), last revised 13 May 2003 (this version, v2)]
Title:Single or combined location measurements of the same parameter without prior probability. (Altern. title, Parametric inference as useful illusion; Part 1)
View PDFAbstract: Motivation. This version is based solely on the calculus of probability, excluding any statistical principle. "Location measurement" means the pdf of the error is known. When the datum is obtained, intuition suggests something like a pdf for the parameter; here we attempt a critical examination of its meaning.
Summary. In default of prior probability the parameter is not defined as a random variable, hence there can be no genuine prior-free parametric inference. Nevertheless prior-free predictive inference regarding any future datum is generated directly from the datum of a location measurement. Such inference turns out as if obtained from a certain pdf ("fiducial") indirectly associated with the parameter. This false pdf can expedite predictive inference, but is inappropriate in the analysis of combined measurements (unless they all are location measurements of the same parameter). Also it has the same distribution as the ostensible Bayesian posterior from a uniform "prior". However, if any of these spurious entities is admitted in the analysis, inconsistent results follow. When we combine measurements, we find that the quantisation errors, inevitable in data recording, must be taken into consideration. These errors cannot be folded into predictive inference in an exact sense; that is, we cannot render a predictive distribution of a future datum except as an approximation.
Keywords: location measurement; combination of observations; parametric inference; predictive inference; prior-free inference; quantisation error; digitisation; frequentist interpretation; the fiducial argument; fiducial probability; pivotal inference; intuitive assessment; prior-free assessment
Submission history
From: George Kahrimanis [view email][v1] Thu, 28 Mar 2002 14:34:32 UTC (19 KB)
[v2] Tue, 13 May 2003 19:43:02 UTC (32 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.