Informal likelihood measures in model assessment: Theoretic development and investigation

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Informal likelihood measures in model assessment: Theoretic development and investigation. / Smith, Paul ; Beven, Keith J.; Tawn, Jonathan A.
In: Advances in Water Resources, Vol. 31, No. 8, 08.2008, p. 1087-1100.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Smith, P , Beven, KJ & Tawn, JA 2008, 'Informal likelihood measures in model assessment: Theoretic development and investigation', Advances in Water Resources, vol. 31, no. 8, pp. 1087-1100. https://doi.org/10.1016/j.advwatres.2008.04.012

APA

Smith, P., Beven, K. J., & Tawn, J. A. (2008). Informal likelihood measures in model assessment: Theoretic development and investigation. Advances in Water Resources, 31(8), 1087-1100. https://doi.org/10.1016/j.advwatres.2008.04.012

Vancouver

Smith P , Beven KJ , Tawn JA. Informal likelihood measures in model assessment: Theoretic development and investigation. Advances in Water Resources. 2008 Aug;31(8):1087-1100. doi: 10.1016/j.advwatres.2008.04.012

Author

Smith, Paul ; Beven, Keith J. ; Tawn, Jonathan A. / Informal likelihood measures in model assessment: Theoretic development and investigation. In: Advances in Water Resources. 2008 ; Vol. 31, No. 8. pp. 1087-1100.

Bibtex

@article{2de33534584d4ed08327cf24f39ca57a,

title = "Informal likelihood measures in model assessment: Theoretic development and investigation",

abstract = "Within hydrology performance criteria such as the Nash-Sutcliffe efficiency have been used to condition the parameter space of a model. Their use is motivated by the fact that the stochastic error series between a model output and corresponding observations is the result of the composite effect of multiple error sources which cannot be described, even in form, a priori. This paper formalises the use of such performance criteria within a Bayesian framework, such as Generalised Likelihood Uncertainty Estimation (GLUE), by introducing the concept of informal Likelihoods. Informal Likelihoods are used to characterise desirable features in the relationship between the model output and corresponding observed data. A number of common performance criteria are considered as Informal Likelihoods. Analytical results and a simulation indicate all of the performance criteria considered as Informal Likelihoods in this paper have one or more properties which may be considered undesirable, but may perform no less well in conditioning model parameters than formal likelihoods for which the assumptions are only mildly incorrect. (C) 2008 Elsevier Ltd. All rights reserved.",

keywords = "GLUE, Bayes theorem, Performance criteria, Informal likelihood",

author = "Paul Smith and Beven, {Keith J.} and Tawn, {Jonathan A.}",

year = "2008",

month = aug,

doi = "10.1016/j.advwatres.2008.04.012",

language = "English",

volume = "31",

pages = "1087--1100",

journal = "Advances in Water Resources",

publisher = "Elsevier Limited",

number = "8",

}

RIS

TY - JOUR

T1 - Informal likelihood measures in model assessment: Theoretic development and investigation

AU - Smith, Paul

AU - Beven, Keith J.

AU - Tawn, Jonathan A.

PY - 2008/8

Y1 - 2008/8

N2 - Within hydrology performance criteria such as the Nash-Sutcliffe efficiency have been used to condition the parameter space of a model. Their use is motivated by the fact that the stochastic error series between a model output and corresponding observations is the result of the composite effect of multiple error sources which cannot be described, even in form, a priori. This paper formalises the use of such performance criteria within a Bayesian framework, such as Generalised Likelihood Uncertainty Estimation (GLUE), by introducing the concept of informal Likelihoods. Informal Likelihoods are used to characterise desirable features in the relationship between the model output and corresponding observed data. A number of common performance criteria are considered as Informal Likelihoods. Analytical results and a simulation indicate all of the performance criteria considered as Informal Likelihoods in this paper have one or more properties which may be considered undesirable, but may perform no less well in conditioning model parameters than formal likelihoods for which the assumptions are only mildly incorrect. (C) 2008 Elsevier Ltd. All rights reserved.

AB - Within hydrology performance criteria such as the Nash-Sutcliffe efficiency have been used to condition the parameter space of a model. Their use is motivated by the fact that the stochastic error series between a model output and corresponding observations is the result of the composite effect of multiple error sources which cannot be described, even in form, a priori. This paper formalises the use of such performance criteria within a Bayesian framework, such as Generalised Likelihood Uncertainty Estimation (GLUE), by introducing the concept of informal Likelihoods. Informal Likelihoods are used to characterise desirable features in the relationship between the model output and corresponding observed data. A number of common performance criteria are considered as Informal Likelihoods. Analytical results and a simulation indicate all of the performance criteria considered as Informal Likelihoods in this paper have one or more properties which may be considered undesirable, but may perform no less well in conditioning model parameters than formal likelihoods for which the assumptions are only mildly incorrect. (C) 2008 Elsevier Ltd. All rights reserved.

KW - GLUE

KW - Bayes theorem

KW - Performance criteria

KW - Informal likelihood

U2 - 10.1016/j.advwatres.2008.04.012

DO - 10.1016/j.advwatres.2008.04.012

M3 - Journal article

VL - 31

SP - 1087

EP - 1100

JO - Advances in Water Resources

JF - Advances in Water Resources

IS - 8

ER -

Research

Associated organisational unit

Links

Text available via DOI:

Keywords