Estimation and Machine Learning Prediction of Imports of Goods in European Countries in the Period 2010-2019

Estimation and Machine Learning Prediction of Imports of Goods in European Countries in the Period 2010-2019

Abstract

In this article we estimate the “Imports of Goods” in European countries in the period 2010-2019 for 28 countries. We use Panel Data with Fixed Effects, Panel Data with Random Effects, Pooled OLS, WLS. Our results show that “Imports of Goods” is negatively associated with “Private Consumption Expenditure at Current Prices”, “Consumption of Fixed Capital”, and “Gross Domestic Product” and positively associated with “Harmonised Consumer Price Index” and “Gross Operating Surplus: Total Economy”. Finally, we compare a set of predictive models based on different machine learning techniques using RapidMiner, and we find that “Gradient Boosted Trees”, “Random Forest”, and “Decision Tree” are more efficient than “Deep Learning”, “Generalized Linear Model” and “Support Vector Machine”, in the sense of error minimization, to forecast the degree of “Imports of Goods”.

Data di pubblicazione

1 October 2021

Autori

  • Prof. Lucio Laureti
  • Prof. Alberto Costantiello
  • Dr. Angelo Leogrande

Introduction

In this article we propose an estimation of an econometric model oriented to determine the degree of “Imports of Goods” in European Countries in the period 2010-2019. We use data from the European Database Ameco for 28 countries4. Data are analyzed using Panel Data with Fixed Effects, Panel Data with Random Effects, Pooled OLS and WLS. Finally, we propose the application of different algorithm-based machine learning techniques to predict the degree of “Imports of Goods” based on the proposed econometric equation. The role of imports in the international trade can be considered as a secondary topic since trade theory seems to be more oriented towards exports rather than imports. But, as we show in the second paragraph, the role of imports of goods is relevant, especially for low-income countries and developing countries, since it is a signal of rising GDP, increasing income per capita, and a strengthening of domestic demand. On the other side the imports of goods in high and middle-income countries have a different dynamic in respect to low-income countries. In effect in high- and middle-income countries imports of services overcome the imports of goods. Specifically, as we showed in our econometric results in the third paragraph, imports of goods in high-income countries are more associated to inputs of firm’s productivity function. To better introduce the theme,we present a brief synthesis of some of the more relevant theories on international trade.

The idea of absolute advantage in Adam Smith. Adam Smith[1] introduced the idea of absolute advantage in the context of exports i.e.,that countries that have lower costs in producing goods are more able to sell them to other countries in the international trade. Originally, the absolute advantage was based on a unique input i.e. labor cost. The countries able to reduce labor cost was also able to win the competition to export in the context of international trade. Specifically, if a country has no possibility to reduce the cost of production, i.e., the cost of labor, then that country has more probabilities to become an importer of that good rather than an exporter. The differences among the presence of absolute advantages create a classification of countries between importers and exporters. 

Welcome to the Age of Digital Transformation

Contattaci per una consulenza gratuita e per lanciare il tuo business digitale.

JOIN US & PAVE THE WAY FOR YOUR INNOVATION ROADMAP


Pubblicazione Details

Corresponding author: Dr. Angelo Leogrande can be contacted at: leogrande.cultore@lum.it