Dissecting the Complexity of Regional STI Policy: A Machine Learning Analysis of Factors Influencing Regional Patent Production

Puntos clave

• The findings indicate that patent production is influenced by nonlinear interactions among innovation factors, with varying impacts across regions.
• Key evidence shows that the optimal level of collaboration and industry diversity thresholds significantly affect patent registration rates.
• The study employs panel regression analysis alongside machine learning methods to investigate the multifaceted relationships at play.
• These insights highlight the need for STI policies to move beyond linear approaches, focusing on combinatory and conditional strategies.

Resumen