HOW OVERCONFIDENCE, BIAS, AND HERDING AFFECT TRADING DECISIONS: THE MODERATING ROLE OF ACCOUNTING KNOWLEDGE

Abstract

This study examines the influence of Overconfidence Bias and Herding Behavior on stock trading decisions among retail traders on the Indonesia Stock Exchange, with Accounting Information Knowledge as a moderating variable. The study is motivated by the rapid surge in retail investor participation reaching 20.32 million by the end of 2025, with 52.59% under the age of 30 dominating 52-77% of daily transactions, yet exhibiting irrational trading patterns characterized by panic selling, overtrading, and portfolio returns lagging 5% below the benchmark index. The sample consists of fifty-one retail traders who actively trade on the Indonesia Stock Exchange, selected through purposive sampling. Structural Equation Modeling-Partial Least Squares was employed to test the proposed hypotheses. The results indicate that Overconfidence Bias has a positive and significant effect on Trading Decisions, suggesting that higher overconfidence leads to more aggressive and impulsive trading decisions. Herding Behavior has a positive and significant effect on Trading Decisions, indicating that retail traders tend to make market trends, community opinion, and social media sentiment the primary basis for decision-making. However, Accounting Information Knowledge does not moderate the relationship between Overconfidence Bias and Trading Decisions, nor between Herding Behavior and Trading Decisions, indicating a gap between textual understanding of financial information and analytical application in real-time decision-making. The findings suggest that psychological biases dominate retail trading behavior while accounting knowledge remains passive and unable to function as a cognitive filter, highlighting the need for literacy programs that develop analytical skills rather than merely transferring knowledge.