Cheap talk and costly consequences

Abstract

In this thesis, we use data on political interactions between country pairs to predict changes in trade. We implemented and applied a new feature selection algorithm called Boruta to build a compact set of predictor variables for this task. After finding a consistent set of features we used a Random Forest Classifier to predict bilateral changes in trade between 1998-2014. To better understand the contribution of each of the predictor variables used in the model we employ three different methods for calculating feature importance. Our results suggest that political and diplomatic interaction at at least as important (if not more) as distance and for predicting changes in trade.