Modeling Investor Behavior in Margin Trading Through Extended TAM and Big Data Analytics: A Systematic Literature Review

The rapid advancement of financial technology (FinTech) has significantly expanded retail participation in capital markets, particularly through margin trading facilities that enable leveraged transactions. In the Indonesian context, this expansion is further accelerated by regulatory developments, such as POJK No. 6 of 2024, which democratize accessibility but simultaneously create a paradox between increased market participation and heightened financial vulnerability due to limited investor financial literacy. This study employs a systematic literature review (SLR) based on PRISMA guidelines across major academic databases to synthesize existing research on investor behavior in margin trading, emphasizing the integration of behavioral finance perspectives with Big Data analytics to process the massive volume of digital footprints generated by these platforms. The results indicate that prior studies predominantly rely on variable-centric models, such as the classical Technology Acceptance Model (TAM), which assume homogeneous behavior and fail to capture the unobserved heterogeneity of investor profiles. To address this limitation, recent literature highlights the relevance of an Extended TAM incorporating trust, perceived risk, and financial literacy. Furthermore, the growing adoption of Big Data analytics, including advanced clustering techniques, provides deeper insights into behavioral discrepancies by mining large-scale historical transaction data. Accordingly, this review advocates for hybrid sequential analytical approaches integrating structural equation modeling (SEM) with data-driven segmentation to generate more precise investor profiles. Ultimately, this study establishes a comprehensive theoretical foundation for future empirical research and offers practical implications for designing adaptive, risk-aware trading platforms.