Explications et équivalence observationnelle: quelques questions à partir d’un modèle économique simple du comportement

Etienne Billette de Villemeur

Post-Print from HAL

Abstract: Assume a « perfect explanation » in that there is a causal model that associates to each set of parameters a unique consequence. Suppose furthermore that the explanation is precise and with no ambiguity in that different outcomes are associated with different parameters – and vice-versa. Should one be satisfied? We show by the means of a simple example that several « perfect explanation » may co-exist. More generally, data need not to be scare, or observations imperfect for « observationally equivalent » explanations to exist. Explanations with an identical predictive power cannot be distinguished on the basis of the sole observations they explain. This indistinguishability raises questions however on the status to be given to their (a priori different) implications. In any case, no matter the amount of data, observations provide only a partial ranking of concurrent explanations.

Date: 2023-05-16
References: Add references at CitEc
Citations:

Published in Bulletin d’histoire et d’épistémologie des sciences de la vie , 2023, 30 (1), pp.35-45. ⟨10.3917/bhesv.301.0035⟩

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-04538521

DOI: 10.3917/bhesv.301.0035

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().