Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Hybrid elicitation and quantile-parametrized likelihood

Perepolkin, Dmytro LU orcid ; Goodrich, Benjamin and Sahlin, Ullrika LU orcid (2023)
Abstract
This paper extends the application of quantile-based Bayesian inference to probability distributions defined in terms of quantiles of observable quantities. Quantile-parameterized distributions are characterized by high shape flexibility and parameter interpretability, making them useful for eliciting information about observables. To encode uncertainty in the quantiles elicited from experts, we propose a Bayesian model based on the metalog distribution and a variant of the Dirichlet prior. We discuss the resulting hybrid expert elicitation protocol, which aims to characterize uncertainty in parameters by asking questions about observable quantities. We also compare and contrast this approach with parametric and predictive elicitation... (More)
This paper extends the application of quantile-based Bayesian inference to probability distributions defined in terms of quantiles of observable quantities. Quantile-parameterized distributions are characterized by high shape flexibility and parameter interpretability, making them useful for eliciting information about observables. To encode uncertainty in the quantiles elicited from experts, we propose a Bayesian model based on the metalog distribution and a variant of the Dirichlet prior. We discuss the resulting hybrid expert elicitation protocol, which aims to characterize uncertainty in parameters by asking questions about observable quantities. We also compare and contrast this approach with parametric and predictive elicitation methods. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Working paper/Preprint
publication status
published
subject
pages
27 pages
publisher
OSF
DOI
10.31219/osf.io/paby6
language
English
LU publication?
yes
id
e71c9579-4089-454b-b9c8-7871bfb070e7
date added to LUP
2021-10-13 13:52:31
date last changed
2024-06-11 11:00:59
@misc{e71c9579-4089-454b-b9c8-7871bfb070e7,
  abstract     = {{This paper extends the application of quantile-based Bayesian inference to probability distributions defined in terms of quantiles of observable quantities. Quantile-parameterized distributions are characterized by high shape flexibility and parameter interpretability, making them useful for eliciting information about observables. To encode uncertainty in the quantiles elicited from experts, we propose a Bayesian model based on the metalog distribution and a variant of the Dirichlet prior. We discuss the resulting hybrid expert elicitation protocol, which aims to characterize uncertainty in parameters by asking questions about observable quantities. We also compare and contrast this approach with parametric and predictive elicitation methods.}},
  author       = {{Perepolkin, Dmytro and Goodrich, Benjamin and Sahlin, Ullrika}},
  language     = {{eng}},
  month        = {{10}},
  note         = {{Preprint}},
  publisher    = {{OSF}},
  title        = {{Hybrid elicitation and quantile-parametrized likelihood}},
  url          = {{http://dx.doi.org/10.31219/osf.io/paby6}},
  doi          = {{10.31219/osf.io/paby6}},
  year         = {{2023}},
}