Precluding rare outcomes by predicting their absence

Eric W. Schoon, David Melamed, Ronald L. Breiger, Eunsung Yoon, Christopher Kleps

Research output: Contribution to journalArticle

Abstract

Forecasting extremely rare events is a pressing problem, but efforts to model such outcomes are often limited by the presence of multiple causes within classes of events, insufficient observations of the outcome to assess fit, and biased estimates due to insufficient observations of the outcome. We introduce a novel approach for analyzing rare event data that addresses these challenges by turning attention to the conditions under which rare outcomes do not occur. We detail how configurational methods can be used to identify conditions or sets of conditions that would preclude the occurrence of a rare outcome. Results from Monte Carlo experiments show that our approach can be used to systematically preclude up to 78.6% of observations, and application to ground-truth data coupled with a bootstrap inferential test illustrates how our approach can also yield novel substantive insights that are obscured by standard statistical analyses.

Original languageEnglish (US)
Article numbere0223239
JournalPloS one
Volume14
Issue number10
DOIs
StatePublished - Jan 1 2019

Fingerprint

pressing
Experiments
testing
methodology

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • General

Cite this

Schoon, E. W., Melamed, D., Breiger, R. L., Yoon, E., & Kleps, C. (2019). Precluding rare outcomes by predicting their absence. PloS one, 14(10), [e0223239]. https://doi.org/10.1371/journal.pone.0223239

Precluding rare outcomes by predicting their absence. / Schoon, Eric W.; Melamed, David; Breiger, Ronald L.; Yoon, Eunsung; Kleps, Christopher.

In: PloS one, Vol. 14, No. 10, e0223239, 01.01.2019.

Research output: Contribution to journalArticle

Schoon, EW, Melamed, D, Breiger, RL, Yoon, E & Kleps, C 2019, 'Precluding rare outcomes by predicting their absence', PloS one, vol. 14, no. 10, e0223239. https://doi.org/10.1371/journal.pone.0223239
Schoon, Eric W. ; Melamed, David ; Breiger, Ronald L. ; Yoon, Eunsung ; Kleps, Christopher. / Precluding rare outcomes by predicting their absence. In: PloS one. 2019 ; Vol. 14, No. 10.
@article{5ca54923b8cb4ed4a094d8852abee318,
title = "Precluding rare outcomes by predicting their absence",
abstract = "Forecasting extremely rare events is a pressing problem, but efforts to model such outcomes are often limited by the presence of multiple causes within classes of events, insufficient observations of the outcome to assess fit, and biased estimates due to insufficient observations of the outcome. We introduce a novel approach for analyzing rare event data that addresses these challenges by turning attention to the conditions under which rare outcomes do not occur. We detail how configurational methods can be used to identify conditions or sets of conditions that would preclude the occurrence of a rare outcome. Results from Monte Carlo experiments show that our approach can be used to systematically preclude up to 78.6{\%} of observations, and application to ground-truth data coupled with a bootstrap inferential test illustrates how our approach can also yield novel substantive insights that are obscured by standard statistical analyses.",
author = "Schoon, {Eric W.} and David Melamed and Breiger, {Ronald L.} and Eunsung Yoon and Christopher Kleps",
year = "2019",
month = "1",
day = "1",
doi = "10.1371/journal.pone.0223239",
language = "English (US)",
volume = "14",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "10",

}

TY - JOUR

T1 - Precluding rare outcomes by predicting their absence

AU - Schoon, Eric W.

AU - Melamed, David

AU - Breiger, Ronald L.

AU - Yoon, Eunsung

AU - Kleps, Christopher

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Forecasting extremely rare events is a pressing problem, but efforts to model such outcomes are often limited by the presence of multiple causes within classes of events, insufficient observations of the outcome to assess fit, and biased estimates due to insufficient observations of the outcome. We introduce a novel approach for analyzing rare event data that addresses these challenges by turning attention to the conditions under which rare outcomes do not occur. We detail how configurational methods can be used to identify conditions or sets of conditions that would preclude the occurrence of a rare outcome. Results from Monte Carlo experiments show that our approach can be used to systematically preclude up to 78.6% of observations, and application to ground-truth data coupled with a bootstrap inferential test illustrates how our approach can also yield novel substantive insights that are obscured by standard statistical analyses.

AB - Forecasting extremely rare events is a pressing problem, but efforts to model such outcomes are often limited by the presence of multiple causes within classes of events, insufficient observations of the outcome to assess fit, and biased estimates due to insufficient observations of the outcome. We introduce a novel approach for analyzing rare event data that addresses these challenges by turning attention to the conditions under which rare outcomes do not occur. We detail how configurational methods can be used to identify conditions or sets of conditions that would preclude the occurrence of a rare outcome. Results from Monte Carlo experiments show that our approach can be used to systematically preclude up to 78.6% of observations, and application to ground-truth data coupled with a bootstrap inferential test illustrates how our approach can also yield novel substantive insights that are obscured by standard statistical analyses.

UR - http://www.scopus.com/inward/record.url?scp=85073097016&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073097016&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0223239

DO - 10.1371/journal.pone.0223239

M3 - Article

C2 - 31600272

AN - SCOPUS:85073097016

VL - 14

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 10

M1 - e0223239

ER -