Random forest based analysis of SME growth under data scarcity in the GCC region

This study examines the drivers of small and medium-sized enterprise (SME) growth in Saudi Arabia within the context of digital transformation and artificial intelligence (AI) readiness. The research is grounded in the Resource-Based View, Dynamic Capabilities, and Institutional Theory. The study develops a quarterly dataset (2018–2022) that integrates SME growth indicators, macroeconomic variables, and pragmatic proxies for digital ecosystem development, including venture capital flows and digital payment infrastructure. A Random Forest model is employed as an exploratory analytical framework to capture non-linear relationships and lagged dynamics under data constraints, with rank-stability and interpolation sensitivity diagnostics used to enhance robustness in a small-sample setting. The results indicate that venture capital deepening and digital payment infrastructure emerge as the most stable correlates of SME growth, while lagged SME performance remains relevant but not dominant. By explicitly positioning proxy-based measurement and machine learning as descriptive tools in data-scarce environments, the study contributes to SME research in emerging economies. The study offers policy and managerially relevant insights into the institutional and ecosystem conditions supporting SME resilience and growth.