The function leans on attributes of the data that are provided by the create_dyadyear()
or
create_stateyear()
function. Make sure that function (or data created by that function) appear at the top
of the proverbial pipe. Users will also want to note that the underlying function access two different data sets. It appears that the
results published in the International Studies Quarterly used Correlates of War classification, but a follow-up repository on Github
uses Gleditsch-Ward classification. The extent to which these estimates are generated by simulation, it does mean the estimates will be slightly
different across both data sets even for common observations (e.g. the United States in 1816).
Because these are large nominal numbers, the estimates have been log-transformed. Users can always exponentiate
these if they choose. Researchers can use these data to construct reasonable estimates of surplus GDP per capita, but must
exponentiate the underlying variables before doing this.