Probabilidad inicial y éxito probabilístico
Análisis Filosófico 29 (1):39-71 (2009)
Abstract
Una cuestión controvertida en la teoría bayesiana de la confirmación es el estatus de las probabilidades iniciales. Aunque la tendencia dominante entre los bayesianos es considerar que la única constricción legítima sobre los valores de dichas probabilidades es la consistencia formal con los teoremas de la teoría matemática de la probabilidad, otros autores -partidarios de lo que se ha dado en llamar "bayesianismo objetivo"- defienden la conveniencia de restricciones adicionales. Mi propuesta, en el marco del bayesianismo objetivo, recoge una sugerencia de Wesley Salmon. No obstante, mientras que para Salmon las probabilidades iniciales se justifican en tanto incorporan la frecuencia de éxito de las hipótesis del pasado, el factor decisivo en mi opinión es el éxito probabilístico de las hipótesis del pasado. Así, la asignación de una probabilidad inicial a una hipótesis h será correcta si refleja el éxito probabilístico de hipótesis del pasado semejantes a h. En esta línea, propongo una regla para fijar las probabilidades iniciales de las hipótesis, y acabo discutiendo algunas objeciones generales en contra de mi planteamiento. One of the most controversial issues for Bayesian theory of confirmation concerns the status of prior probabilities. Orthodox Bayesianism claims that logical consistency with the theorems of mathematical theory of probability is the only constraint on prior probabilities, while some authors claim for further requirements. I agree with the latter standpoint. My proposal is inspired by Wesley Salmon's frequentist justification of prior probabilities. According to him, prior probabilities are the best assessments of past scientific hypotheses' success. I think that historical information is relevant here, but I propose to focus on a particular sort of success -i.e., "probabilistic success". I claim that a particular prior for h is correct insofar as it reflects the probabilistic success of past scientific hypotheses similar to h. I also suggest a rule to set the priors for contemporary hypotheses. The closing paragraphs are devoted to some general objections against this proposalAuthor's Profile
My notes
Similar books and articles
Confirmación hipotético-deductiva y confirmación bayesiana.Alejandro Cassini - 2003 - Análisis Filosófico 23 (1):41-84.
Optimum Inductive Methods: A Study in Inductive Probability, Bayesian Statistics, and Verisimilitude.Roberto Festa - 1993 - Dordrecht, Netherland: Kluwer Academic Publishers: Dordrecht.
Bayesian decision theory, subjective and objective probabilities, and acceptance of empirical hypotheses.John C. Harsanyi - 1983 - Synthese 57 (3):341 - 365.
On Ratio Measures of Confirmation: Critical Remarks on Zalabardo’s Argument for the Likelihood-Ratio Measure.Valeriano Iranzo & Ignacio Martínez de Lejarza - 2013 - Journal for General Philosophy of Science / Zeitschrift für Allgemeine Wissenschaftstheorie 44 (1):193-200.
The status of prior probabilities in statistical explanation.Wesley C. Salmon - 1965 - Philosophy of Science 32 (2):137-146.
Éxito, razón Y cambio en fisíca: Un enfoque instrumental en teoría de la ciencia.Valeriano Iranzo - 2006 - Theoria 21 (1):113-114.
Constraining prior probabilities of phylogenetic trees.Bengt Autzen - 2011 - Biology and Philosophy 26 (4):567-581.
Subjective Probabilities as Basis for Scientific Reasoning?Franz Huber - 2005 - British Journal for the Philosophy of Science 56 (1):101-116.
The prior probabilities of phylogenetic trees.Joel D. Velasco - 2008 - Biology and Philosophy 23 (4):455-473.
Analytics
Added to PP
2013-10-03
Downloads
30 (#390,986)
6 months
1 (#447,139)
2013-10-03
Downloads
30 (#390,986)
6 months
1 (#447,139)
Historical graph of downloads
Author's Profile
References found in this work
Bayes or Bust?: A Critical Examination of Bayesian Confirmation Theory.John Earman - 1992 - Bradford.
A subjectivist’s guide to objective chance.David K. Lewis - 1980 - In Richard C. Jeffrey (ed.), Studies in Inductive Logic and Probability, Volume II. Berkeley: University of California Press. pp. 263-293.
Bayes or Bust?: A Critical Examination of Bayesian Confirmation Theory.John Earman - 1992 - MIT Press.
The Foundations of Scientific Inference.Wesley C. Salmon - 1967 - Pittsburgh, PA, USA: University of Pittsburgh Pre.