Forecasting the macrolevel determinants of entrepreneurial opportunities using artificial intelligence models

To date, entrepreneurship researchers have tended to avoid state-of-the-art artificial intelligence techniques; in this paper, we fill that gap. Based on eclectic entrepreneurship theory, we present an original work that uses artificial intelligence to forecast the macrolevel determinants of entrepreneurial opportunity. Modern artificial intelligence could open new areas for future research opportunities in entrepreneurship and help close the gap between theory and practice. Our empirical analysis offers two major results by using a panel dataset of 149 countries covering 2007–2018 and six machine-learning models. First, entrepreneurs prefer to exploit opportunities in countries with stable economic governance that provide high education standards, health, social capital, and a safe, natural environment. Second, CatBoost regression performs better in predicting entrepreneurial opportunity compared to linear regression and more advanced machine-learning models. Recommendations for policy-makers and managers and directions for future studies are also discussed.