Accounting information and stock returns in Vietnam securities market: Machine learning approach

Abstract

This paper studies the relationship between accounting information reflected in financial statements and stock return in Vietnam Stock Market. The authors propose a research model to define the relationship. Besides, machine learning algorithms are used in researching and forecasting, with data obtained from observed firms during the period from 2009 to 2020. Research results show that gradient boosting algorithm has the best self-reporting performance and finan-cial ratios also have great impact on stock returns, including operating income growth, stock earnings volatility, dividend yield, earnings before tax-to-equity ratio, cash holding ratio, and accrual quality. Based on the research results, the authors make some recommendations for investors, firms, and policy makers.