The reconciliation of multiple conflicting estimates: Entropy-based and axiomatic approaches

João F.D. Rodrigues, Michael L. Lahr

Research output: Contribution to journalArticle

Abstract

When working with economic accounts it may occur that multiple estimates of a single datum exist, with different degrees of uncertainty or data quality. This paper addresses the problem of defining a method that can reconcile conflicting estimates, given best guess and uncertainty values. We proceeded from first principles, using two different routes. First, under an entropy-based approach, the data reconciliation problem is addressed as a particular case of a wider data balancing problem, and an alternative setting is found in which the multiple estimates are replaced by a single one. Afterwards, under an axiomatic approach, a set of properties is defined, which characterizes the ideal data reconciliation method. Under both approaches, the conclusion is that the formula for the reconciliation of best guesses is a weighted arithmetic average, with the inverse of uncertainties as weights, and that the formula for the reconciliation of uncertainties is a harmonic average.

Original languageEnglish (US)
Article number815
JournalEntropy
Volume20
Issue number11
DOIs
StatePublished - Nov 1 2018

Fingerprint

entropy
estimates
economics
routes
harmonics

All Science Journal Classification (ASJC) codes

  • Physics and Astronomy(all)

Keywords

  • Axiomatix approach
  • Conflicting estimates
  • Economic accounts
  • Entropy-based approach
  • Uncertainty modelling

Cite this

@article{e052f79359c4444fb64a8a7a33426e28,
title = "The reconciliation of multiple conflicting estimates: Entropy-based and axiomatic approaches",
abstract = "When working with economic accounts it may occur that multiple estimates of a single datum exist, with different degrees of uncertainty or data quality. This paper addresses the problem of defining a method that can reconcile conflicting estimates, given best guess and uncertainty values. We proceeded from first principles, using two different routes. First, under an entropy-based approach, the data reconciliation problem is addressed as a particular case of a wider data balancing problem, and an alternative setting is found in which the multiple estimates are replaced by a single one. Afterwards, under an axiomatic approach, a set of properties is defined, which characterizes the ideal data reconciliation method. Under both approaches, the conclusion is that the formula for the reconciliation of best guesses is a weighted arithmetic average, with the inverse of uncertainties as weights, and that the formula for the reconciliation of uncertainties is a harmonic average.",
keywords = "Axiomatix approach, Conflicting estimates, Economic accounts, Entropy-based approach, Uncertainty modelling",
author = "Rodrigues, {Jo{\~a}o F.D.} and Lahr, {Michael L.}",
year = "2018",
month = "11",
day = "1",
doi = "https://doi.org/10.3390/e20110815",
language = "English (US)",
volume = "20",
journal = "Entropy",
issn = "1099-4300",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "11",

}

The reconciliation of multiple conflicting estimates : Entropy-based and axiomatic approaches. / Rodrigues, João F.D.; Lahr, Michael L.

In: Entropy, Vol. 20, No. 11, 815, 01.11.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The reconciliation of multiple conflicting estimates

T2 - Entropy-based and axiomatic approaches

AU - Rodrigues, João F.D.

AU - Lahr, Michael L.

PY - 2018/11/1

Y1 - 2018/11/1

N2 - When working with economic accounts it may occur that multiple estimates of a single datum exist, with different degrees of uncertainty or data quality. This paper addresses the problem of defining a method that can reconcile conflicting estimates, given best guess and uncertainty values. We proceeded from first principles, using two different routes. First, under an entropy-based approach, the data reconciliation problem is addressed as a particular case of a wider data balancing problem, and an alternative setting is found in which the multiple estimates are replaced by a single one. Afterwards, under an axiomatic approach, a set of properties is defined, which characterizes the ideal data reconciliation method. Under both approaches, the conclusion is that the formula for the reconciliation of best guesses is a weighted arithmetic average, with the inverse of uncertainties as weights, and that the formula for the reconciliation of uncertainties is a harmonic average.

AB - When working with economic accounts it may occur that multiple estimates of a single datum exist, with different degrees of uncertainty or data quality. This paper addresses the problem of defining a method that can reconcile conflicting estimates, given best guess and uncertainty values. We proceeded from first principles, using two different routes. First, under an entropy-based approach, the data reconciliation problem is addressed as a particular case of a wider data balancing problem, and an alternative setting is found in which the multiple estimates are replaced by a single one. Afterwards, under an axiomatic approach, a set of properties is defined, which characterizes the ideal data reconciliation method. Under both approaches, the conclusion is that the formula for the reconciliation of best guesses is a weighted arithmetic average, with the inverse of uncertainties as weights, and that the formula for the reconciliation of uncertainties is a harmonic average.

KW - Axiomatix approach

KW - Conflicting estimates

KW - Economic accounts

KW - Entropy-based approach

KW - Uncertainty modelling

UR - http://www.scopus.com/inward/record.url?scp=85057010883&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057010883&partnerID=8YFLogxK

U2 - https://doi.org/10.3390/e20110815

DO - https://doi.org/10.3390/e20110815

M3 - Article

VL - 20

JO - Entropy

JF - Entropy

SN - 1099-4300

IS - 11

M1 - 815

ER -