Theory and evidence in international conflict

A response to de marchi, gelpi, and grynaviski

Nathaniel Beck, Gary King, Langche Zeng

Research output: Contribution to journalReview article

Abstract

In this article, we show that de Marchi, Gelpi, and Grynaviski's substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model.

Original languageEnglish (US)
Pages (from-to)379-389
Number of pages11
JournalAmerican Political Science Review
Volume98
Issue number2
DOIs
StatePublished - May 2004

Fingerprint

international conflict
evidence
neural network
democracy

ASJC Scopus subject areas

  • Sociology and Political Science

Cite this

Theory and evidence in international conflict : A response to de marchi, gelpi, and grynaviski. / Beck, Nathaniel; King, Gary; Zeng, Langche.

In: American Political Science Review, Vol. 98, No. 2, 05.2004, p. 379-389.

Research output: Contribution to journalReview article

@article{719eb0a2c3604bb8b8777766c8d7aa6d,
title = "Theory and evidence in international conflict: A response to de marchi, gelpi, and grynaviski",
abstract = "In this article, we show that de Marchi, Gelpi, and Grynaviski's substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model.",
author = "Nathaniel Beck and Gary King and Langche Zeng",
year = "2004",
month = "5",
doi = "10.1017/S0003055404001212",
language = "English (US)",
volume = "98",
pages = "379--389",
journal = "American Political Science Review",
issn = "0003-0554",
publisher = "Cambridge University Press",
number = "2",

}

TY - JOUR

T1 - Theory and evidence in international conflict

T2 - A response to de marchi, gelpi, and grynaviski

AU - Beck, Nathaniel

AU - King, Gary

AU - Zeng, Langche

PY - 2004/5

Y1 - 2004/5

N2 - In this article, we show that de Marchi, Gelpi, and Grynaviski's substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model.

AB - In this article, we show that de Marchi, Gelpi, and Grynaviski's substantive analyses are fully consistent with our prior theoretical conjecture about international conflict. We note that they also agree with our main methodological point that out-of-sample forecasting performance should be a primary standard used to evaluate international conflict studies. However, we demonstrate that all other methodological conclusions drawn by de Marchi, Gelpi, and Gryanaviski are false. For example, by using the same evaluative criterion for both models, it is easy to see that their claim that properly specified logit models outperform neural network models is incorrect. Finally, we show that flexible neural network models are able to identify important empirical relationships between democracy and conflict that the logit model excludes a priori; this should not be surprising since the logit model is merely a limiting special case of the neural network model.

UR - http://www.scopus.com/inward/record.url?scp=3042806894&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3042806894&partnerID=8YFLogxK

U2 - 10.1017/S0003055404001212

DO - 10.1017/S0003055404001212

M3 - Review article

VL - 98

SP - 379

EP - 389

JO - American Political Science Review

JF - American Political Science Review

SN - 0003-0554

IS - 2

ER -