This file describes script files (STATA do files) and datasets used for replicating estimation results from 
Hegre et al 2013 "Predicting the Future of Armed Conflict, 2010--2050".

All estimations are done using STATA 11.2

Datasets:


estimationdata_withlags.dta
This dataset contains all the variables used for estimating tables A-1 and A-2 in the paper. All variable manipulations have been
done for this dataset

aggregatedglobalpredictions.dta
This dataset contains the aggregated prediction results. I.e. the global annual share of countries in conflict, with percentiles. The dataset contains separate predictions for each of the nine models in the paper (see Figure 10). 


CountryYearResults.dta
This dataset contains the predicted risk of conflict for every country for every year based on the combined model reported in most tables in the article.

Scripts:

estimationreplication.do
This STATA do file contains the commands for estimating the 9 models in tables A-1 and A-2 in the paper. It calls on the dataset
estimationdata_withlags.dta, defines a set of constraints and estimates the models. Stata 11 is needed to run the models. To 
estimate the models on earlier versions of STATA the syntax for the constraints must be changed.

