Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 442 - Contributed Poster Presentations: Section on Statistics and Data Science Education
Type: Contributed
Date/Time: Wednesday, August 10, 2022 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics and Data Science Education
Abstract #320990
Title: Time to War in the Thucydides Trap
Author(s): Jake Tan* and Chris McDaniels
Companies: Wissahickon High School and Wissahickon High School
Keywords: time-to-event; survival analysis; model averaging; AIC

Harvard Kennedy School Professor Graham Allison has published many articles and a book on the Thucydides Trap. The basic idea is that "when one great power threatens to displace another, war is almost always the result". For the 16 cases listed by Professor Allison, we quantified the time from competition engagement to outbreak of war in order to model the time course pattern. First, we fit six parametric survival models to the “time to war” data. Based on the AICc criterion, the exponential model is identified as the best model. Then, we combined all six models using Akaike weights. Based on the combined model, we predict that the risk or hazard of war is about 37% per decade, but in the first decade the risk is slightly elevated to 40%, and in the fifth decade the risk is slightly depressed to 34%. These findings may inform us about the trajectory of current and future great power competitions with Thucydides Trap dynamics.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program