Do Short Exposure and Systematic Risk Exposure Drive Asymmetries in the Disposition Effect?

Battle Count: 81

The paper provides important insights into retail investor behavioral biases that can inform quantitative trading strategies, particularly in leveraged ETF markets. Understanding disposition effect asymmetries between long/short positions and across portfolio states can help design contrarian strategies, improve execution algorithms, and develop behavioral alpha signals. The systematic risk amplification findings are relevant for risk management in leveraged products. However, the paper is primarily descriptive/behavioral rather than prescriptive for trading strategy construction.

The dispositionEffect R package simplifies implementation of the core measures (Count, Total, Value) across narrow, wide, and integrated framing. However, the full integrated framing methodology requires real-time portfolio reconstruction from transaction-level data, which is computationally intensive. The Mann-Whitney U-test comparisons are straightforward. The main complexity lies in data preparation (intraday price series construction, portfolio state tracking) and handling large-scale transaction datasets (~9 million records).

Code is publicly available on CRAN (dispositionEffect package) and GitHub, enabling replication of the methodology. However, the underlying transaction data from Directa brokerage is proprietary and cannot be publicly shared, limiting full reproducibility. The methodology is well-documented and the R package provides standardized computation of all measures.