Machine learning-driven insights for business innovation: Analyzing financial performance in Vietnam’s emerging economy 2010–2024

This paper examines the impact of Vietnam's emerging market conditions on financial performance forecasting for listed firms, where stock market capitalization reached 56 % of Gross Domestic Product (GDP) in 2023. Using a sample of 551 firms across eight industries over the period 2010–2024, we compare machine learning (ML) techniques, specifically Random Forest, with traditional econometric models to predict Return on Assets, drawing on standardized financial data platforms. Unlike prior studies on stock prices or financial distress, our findings indicate that ML enhances predictive accuracy, with the effect channeled through knowledge integration via data platforms. Furthermore, anchored in Resource-Based View (RBV), Transaction Cost Theory (TCT), and Industrial Organization Theory (IOT), we document that firm-specific factors, including operational efficiency and financial stability, exert a stronger influence than macroeconomic conditions. Capital-intensive sectors, such as Industrials, exhibit superior forecasting accuracy compared to consumer sectors, attributable to consistent data availability. This study introduces a novel framework for emerging markets, with Vietnam as a representative case to illustrate applications in similar economies, enabling managers to refine financial strategies, investors to optimize portfolios, and policymakers to enhance digital infrastructure for sustainable economic growth.