Is more better or worse? New empirics on nuclear proliferation and interstate conflict by Random Forests

Abstract

In the literature on nuclear proliferation, some argue that further proliferation decreases interstate conflict, some say that it increases interstate conflict, and others indicate a non-linear relationship between these two factors. However, there has been no systematic empirical investigation on the relationship between nuclear proliferation and a propensity for conflict at the interstate-systemic level. To fill this gap, the current paper uses the machine learning method Random Forests, which can investigate complex non-linear relationships between dependent and independent variables, and which can identify important regressors from a group of all potential regressors in explaining the relationship between nuclear proliferation and the propensity for conflict. The results indicate that, on average, a larger number of nuclear states decrease the systemic propensity for interstate conflict, while the emergence of new nuclear states does not have an important effect. This paper also notes, however, that scholars should investigate other risks of proliferation to assess whether nuclear proliferation is better or worse for international peace and security in general.

Entropy-based discrimination between translated Chinese and original Chinese using data mining techniques

“…Random forests work by generating many decision trees from random samples of the data and selecting the features which reliably best predict the outcomes. Random forests have, like the other classification algorithms, been used across a wide range of fields and for many very distinct tasks, from predicting how sensitive tumors are to drugs [69], to predicting the onset of civil wars [70], to predicting the role of nuclear proliferation on interstate conflict [71].…”

How to Cautiously Uncover the “Black Box” of Machine Learning Models for Legislative Scholars

“…. Last, Random Forests are popular ML tools that have been used elsewhere in political science (Funk, Paul, and Philips Forthcoming; Muchlinski et al 2016; Suzuki 2015), including legislative politics (Bonica 2018), helping ensure Random Forest models employed by legislative scholars would find wider audiences within political science.…”

“…Despite recent articles in political science espousing the benefits of ML models (cf. Montgomery and Olivella 2018), ML model interpretation has typically focused on predictive accuracy rather than visualizing the relationship between the predictors and the outcome (Anastasopoulos and Bertelli 2020; Kaufman, Kraft, and Sen 2019; with some exceptions, Green and Kern 2012; Kim, Alvarez, and Ramirez 2020; Muchlinski et al 2016; Suzuki 2015). This “black‐box” complexity has kept machine learning models from more prevalent usage in legislative studies; we could only find 16 uses of machine learning models in Legislative Studies Quarterly from 2010 to 2020, despite their broader growth in political science.…”

Predicting the issuance of COVID-19 stay-at-home orders in Africa: Using machine learning to develop insight for health policy research

“…Beneficially, the random forest approach still allows for the application of theoretical arguments after the fact. Because of these advantageous properties, random forests are being utilized in economics to study problems of mass appraisal, evaluation, and market prediction [ 8 , 12 , 13 ], and in political science to study civil war onset [ 9 , 14 ], the effect of IMF policy on child poverty rates [ 15 ], nuclear proliferation [ 16 ], and migration [ 17 ].…”