Irregular leadership changes in 2014: Forecasts using ensemble, split-population duration models

Beger, Andreas; Dorff, Cassy; Ward, Michael D.

doi:10.1016/j.ijforecast.2015.01.009

Cited by 28 publications

(24 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We apply several machine learning methods. Since each has its own strengths and drawbacks discussed below, we also take a weighted average of the four using an Ensemble Bayesian Model Average (Beger et al, 2016). Starting with the prior that each algorithm is equally appropriate, we use cross-validation to update our weights based on the accuracy of each model (Montgomery et al, 2012).…”

Section: Machine Learning Algorithmsmentioning

confidence: 99%

The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia

Bazzi

Blair

Blattman

et al. 2022

The Review of Economics and Statistics

View full text Add to dashboard Cite

How feasible is violence early-warning prediction? Columbia and Indonesia have unusually fine-grained data. We assemble two decades of local violent events alongside hundreds of annual risk factors. We attempt to predict violence one year ahead with a range of machine learning techniques. Our models reliably identify persistent, high-violence hot spots. Violence is not simply autoregressive, as detailed histories of disaggregated violence perform best, but socioeconomic data substitute well for these histories. Even with unusually rich data, however, our models poorly predict new outbreaks or escalations of violence. These “best case” scenarios with annual data fall short of workable early-warning systems.

show abstract

Section: Machine Learning Algorithmsmentioning

confidence: 99%

The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia

Bazzi

Blair

Blattman

et al. 2022

The Review of Economics and Statistics

View full text Add to dashboard Cite

show abstract

“…This equifinality -the different political processes that can lead to an ART -complicates the use of more traditional methodological approaches in prediction problems. Indeed, while traditional methods of description, explanation, and inference can perform well at explaining the onset of important political events ex-post, when researchers apply these well-established methods to prediction problems, they tend to perform poorly (Beger et al 2016, Hill & Jones 2014, Schrodt 2014, Soyer & Hogarth 2012, Ward et al 2010. To overcome these problems, social scientists have been borrowing forecasting methods from other fields -machine learning methods from computer science, in particular.…”

Section: Modelingmentioning

confidence: 99%

“…There are a number of forecasting efforts currently underway throughout the international relations and comparative politics research communities. These projects range from predicting the onset of civil and international conflict (Brandt et al 2011, Hegre et al 2019, 2013 and mass killings and atrocities (Goldsmith & Butcher 2018, Goldsmith et al 2013, Woocher et al 2018 to whether a country will experience an irregular leadership change (Beger et al 2016, Ward & Beger 2017 and political instability, in general (Goldstone et al 2010). For the most part, these forecasting efforts focus on estimating the potential risk that a country will experience some form of political violence.…”

Section: Introduction and Executive Summarymentioning

confidence: 99%

Varieties of Forecasts: Predicting Adverse Regime Transitions

Morgan

Beger²,

Glynn

2019

SSRN Journal

Self Cite

View full text Add to dashboard Cite

This article introduces the V-Forecast project, the forecasting intuitive of the Varieties of Democracy (V-Dem) institute. In this the initial year of the V-Forecast project, we provide two-year ahead forecasts of the risk of adverse regime transitions (ARTs) for 169 countries. ARTs are substantial movements of a country's regime towards more authoritarian governance, whether authoritarian reversals in a democracy, or further autocratization in an already nondemocratic country. Examples include Hungary and Poland over the past few years, which are prominent cases in a more widespread and worrying global trend over that effects a significant fraction of the world's population. Yet so far, there has been no public forecasting system for anticipating new ARTs and identifying countries most at risk. We describe an effort that forecasts ARTs-operationalized using the Regimes of the World (RoW) categorization-with an ensemble model that leverages V-Dem and several additional external data sources. Despite being rare events with a roughly four percent baseline chance over any two-year period, in test forecasts the model is able to achieve good accuracy.

show abstract

“…As such, our model shares similarities with the cure survival model, which has been previously used in Political Science to model competing processes of democratic survival (Svolik 2008), or to accommodate heterogeneous mixtures of “at risk” and “not at risk” countries in global analyses of irregular leadership changes (Beger, Dorff, and Ward 2014, 2015; Beger et al. 2017).…”

Section: Introductionmentioning

confidence: 99%

A Bayesian Split Population Survival Model for Duration Data With Misclassified Failure Events

Bagozzi¹,

Joo²,

Kim³

et al. 2019

Polit. Anal.

View full text Add to dashboard Cite

We develop a new Bayesian split population survival model for the analysis of survival data with misclassified event failures. Within political science survival data, right-censored survival cases are often erroneously misclassified as failure cases due to measurement error. Treating these cases as failure events within survival analyses will underestimate the duration of some events. This will bias coefficient estimates, especially in situations where such misclassification is associated with covariates of interest. Our split population survival estimator addresses this challenge by using a system of two equations to explicitly model the misclassification of failure events alongside a parametric survival process of interest. After deriving this model, we use Bayesian estimation via slice sampling to evaluate its performance with simulated data, and in several political science applications. We find that our proposed “misclassified failure” survival model allows researchers to accurately account for misclassified failure events within the contexts of civil war duration and democratic survival.

show abstract

Irregular leadership changes in 2014: Forecasts using ensemble, split-population duration models

Cited by 28 publications

References 67 publications

The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia

The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia

Varieties of Forecasts: Predicting Adverse Regime Transitions

A Bayesian Split Population Survival Model for Duration Data With Misclassified Failure Events

Contact Info

Product

Resources

About