Product Progression: a machine learning approach to forecasting industrial upgrading

TL;DR

This paper introduces a machine learning framework for forecasting industrial upgrading by predicting new product activations, demonstrating that tree-based models outperform benchmarks and other algorithms, with implications for policy and economic development.

Contribution

It provides a systematic evaluation of machine learning models for forecasting product activation, highlighting the effectiveness of tree-based algorithms and cross-validation for economic predictions.

Findings

Tree-based algorithms outperform other models in forecasting

Forecasting accuracy improves with cross-validation excluding the target country

The approach offers policy-relevant measures for industrial upgrading feasibility

Abstract

Economic complexity methods, and in particular relatedness measures, lack a systematic evaluation and comparison framework. We argue that out-of-sample forecast exercises should play this role, and we compare various machine learning models to set the prediction benchmark. We find that the key object to forecast is the activation of new products, and that tree-based algorithms clearly overperform both the quite strong auto-correlation benchmark and the other supervised algorithms. Interestingly, we find that the best results are obtained in a cross-validation setting, when data about the predicted country was excluded from the training set. Our approach has direct policy implications, providing a quantitative and scientifically tested measure of the feasibility of introducing a new product in a given country.