Predicting Exporters with Machine Learning

In this contribution, we exploit machine learning techniques to predict out-of-sample firms' ability export based on the financial accounts of both exporters and non-exporters. Therefore, show how forecasts can be used as exporting scores, i.e., measure distance non-exporters from status. For our purpose, train test various algorithms reports 57,021 manufacturing firms in France 2010-2018. We find that a Bayesian Additive Regression Tree with Missingness Attributes (BART-MIA) performs better than other prediction accuracy up 0.90. Predictions are robust changes definitions presence discontinuous exporters. Eventually, argue scores helpful for trade promotion, credit, assess competitiveness. example, back-of-the-envelope estimates representative firm just below-average needs 44% more cash resources 2.5 times capital expenses reach full