Irregular Leadership Changes in 2014: Forecasts using ensemble, split-population duration models

TL;DR

This paper develops an ensemble of split-population duration models to forecast irregular leadership changes globally, providing actionable predictions for mid-2014 based on historical data and model accuracy.

Contribution

It introduces an innovative ensemble modeling approach combining thematic split-population duration models for predicting irregular leadership changes.

Findings

High-probability countries include Ukraine, Yemen, Egypt, Thailand, Bosnia & Herzegovina.

The ensemble model accurately forecasts leadership change risks for mid-2014.

The approach improves prediction accuracy over single-model methods.

Abstract

We forecast Irregular Leadership Changes (ILC)--unexpected leadership changes in contravention of a state's established laws and conventions--for mid-2014 using predictions generated from an innovative ensemble model that is composed of several split-population duration regression models. This approach uses distinct thematic models, combining them into one aggregate forecast developed on the basis of their predictive accuracy and uniqueness. The data are based on 45 ILCs that occurred from March 2001 through March 2014, with monthly observations for up to 168 countries worldwide. The ensemble model provides forecasts for the period from April to September 2014. Notably, the countries with the highest probability of irregular leadership change in the middle six months of 2014 include the Ukraine, Bosnia & Herzegovina, Yemen, Egypt, and Thailand. The leadership in these countries have exhibited fragility during this forecast window.