Abstract
In a recent paper Hernández and Johnson (1984) have given a procedure based on Bayesian statistical inference for selecting an extreme-value distribution to “best” fit available data. In this note we give an alternative derivation of part of their results. This derivation — based onstructural inference (Fraser 1968) — provides theoretical support for these results that would not be present on the basis of their derivation using Bayesian techniques.
Zusammenfassung
Für die Auswahl einer Extremwertverteilung, die vorgelegten Daten “optimal” angepaßt ist, wurde von Hernández und Johnson (1984) ein Verfahren angegeben, das auf dem Bayes-Prinzip beruht. In dieser Arbeit wird ein anderer Zugang zu einem Teil jener Resultate aufgezeigt. Dieser auf der “structural inference” von Fraser (1968) beruhende Zugang liefert eine zusätzliche theoretische Rechtfertigung jener Verfahren.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Brenner D, Fraser DAS (1982) On the foundations: statistical models and inference. Canadian Journal of Statistics 10:155–161
Fraser DAS (1968) The structure of inference. John Wiley, New York
Hernández F, Johnson RA (1984) Selecting an extreme-value distribution and transforming to a specified distribution. Operations Research 32:715–725
Author information
Authors and Affiliations
Additional information
Supported in part by a Washington State University Faculty Summer Support Award 1984.
Rights and permissions
About this article
Cite this article
Ariyawansa, K.A. On selecting an extreme value distribution. Zeitschrift für Operations Research 32, 95–100 (1988). https://doi.org/10.1007/BF01919184
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01919184