Change Point Analysis of Historical Battle Deaths

Abstract

It has been claimed and disputed that World War II has been followed by a 'long peace': an unprecedented decline of war. We conduct a full change point analysis of welldocumented, publicly available battle deaths data sets, using new techniques that enable the robust detection of changes in the statistical properties of such heavy-tailed data. We first test and calibrate these techniques. We then demonstrate the existence of changes, independent of data presentation, in the early to mid-19th century, as the Congress of Vienna system moved towards its collapse, in the early to mid-20th century, bracketing the World Wars, and in the late 20th century, as the world reconfigured around the end of the Cold War. Our analysis provides a methodology for future investigations and an empirical basis for political and historical discussions.

Statistical sightings of better angels: Analysing the distribution of battle-deaths in interstate conflict over time

“…Figure 1 displays these data, with z i on the log10-scale. The choice of the CoW dataset is motivated by its widespread use (Clauset, 2017(Clauset, , 2018Fagan et al, 2018;Spagat & van Weezel, 2018), which enables comparisons with other approaches. Also, the CoW dataset is considered to be of good quality, despite the issue mentioned above.…”

“…There are several recent contributions with clear connections to our article. Many of these also analyse the CoW interstate conflict dataset (Clauset, 2017(Clauset, , 2018Spagat & van Weezel, 2018;Fagan et al, 2018;Braumoeller, 2019), while Cederman, Warren & Sornette (2011) and Cirillo & Taleb (2016) use datasets with a longer time span (from year 1494, and year 1, respectively). Cirillo & Taleb (2016) and Spagat & van Weezel (2018) normalize the war sizes with respect to world population, while Clauset (2017Clauset ( , 2018 and Cederman, Warren & Sornette (2011) analyse the absolute numbers.…”

“…Cirillo & Taleb (2016) and Spagat & van Weezel (2018) normalize the war sizes with respect to world population, while Clauset (2017Clauset ( , 2018 and Cederman, Warren & Sornette (2011) analyse the absolute numbers. Fagan et al (2018) conduct analyses of both absolute and relative numbers. As expected, analyses using relative war sizes find a clearer decline of war than those focusing on absolute numbers.…”

“…Parametric models within the class of distributions with power law tails are used in Cederman, Warren & Sornette (2011), Cirillo & Taleb (2016) and Clauset (2017Clauset ( , 2018, while Fagan et al (2018) and Spagat & Weezel (2018) use non-parametric approaches. Clauset (2017Clauset ( , 2018 also investigated a certain semi-parametric model.…”

Lewis Fry Richardson: His Intellectual Legacy and Influence in the Social Sciences

“…Until recently scholars have tended to assume that the Second World War is the obvious candidate for a break point into a more peaceful world. However, recent papers by Fagan et al (2018) and Cunen et al (2018) start from an agnostic position on potential break points and use statistical methods to detect convincing ones. Both papers find substantial evidence for a change at 1950 although they identify other candidate break points including 1912 (Fagan et al, 2018) and 1965 (Cunen et al, 2018).…”