Replication Data for: Prediction, Proxies, and Power

Many enduring questions in international relations theory focus on power relations, so it is important that scholars have a good measure of relative power. The standard measure of relative military power, the capability ratio, is barely better than random guessing at pre- dicting militarized dispute outcomes. We use machine learning to build a superior proxy, the Dispute Outcome Expectations score, from the same underlying data. Our measure is an order of magnitude better than the capability ratio at predicting dispute outcomes. We replicate Reed et al. (2008) and find, contrary to the original conclusions, that the probability of conflict is always highest when the state with the least benefits has a preponderance of power. In replications of 18 other dyadic analyses that use power as a control, we find that replacing the standard measure with DOE scores usually improves both in-sample and out-of-sample goodness of fit. Note:This analysis involves many layers of computation: multiple imputation of the underlying data, creation of an ensemble of machine learning models on the imputed datasets, predictions from that ensemble, and replications of previous studies using those predictions. Our replication code sets seeds in any script where random numbers are drawn, and runs in a Docker environment to ensure identical package versions across machines. Nevertheless, because of differences in machine precision and floating point computations across CPUs, the replication code may not produce results identical to those in the paper. Any differences should be small in magnitude and should not affect any substantive conclusions of the analysis.