The Empirical Analysis of Investor Behavioral Biases and Their Predictive Impact on Total Bias

Abstract

This study examines the prevalence and predictive impact of investor behavioural biases on aggregate decision-making distortions, referred to as Total Bias. A quantitative research design was adopted, and primary data were collected from 360 investors using a structured questionnaire and a combination of convenience and purposive sampling. The study investigated thirteen behavioural biases, including overconfidence, loss aversion, anchoring, herding, availability, regret aversion, hindsight, cognitive aversion and self-attribution. Data were analysed using one-sample t-tests, Pearson correlation analysis and stepwise multiple regression. The findings revealed that most behavioural biases had significantly lower observed mean scores than the expected benchmark values, indicating systematic deviations from rational decision-making. Correlation analysis showed that herding bias (R = .599, p < .001), regret aversion bias (R = .568, p < .001) and cognitive aversion bias (R = .536, p < .001) had the strongest positive relationships with Total Bias. Stepwise regression identified herding bias as the most influential predictor, followed by self-attribution, cognitive aversion, anchoring and regret aversion biases. The inclusion of successive biases substantially improved the explanatory power of the model, culminating in a near-perfect fit (R² = 1.00). The study highlights the interconnected influence of cognitive and emotional biases on investor behaviour and contributes to the growing behavioural finance literature. The findings offer practical implications for investors, financial advisers, portfolio managers and policymakers by emphasising the need for investor education, behavioural awareness programmes and bias-mitigation strategies to improve investment decision-making and promote market efficiency.