Revisiting Boehmer et al. (2021): Recent Period, Alternative Method, Different Conclusions
With David Ardia and Tolga Cenesizoglu. Published in Financial Markets and Portfolio Management, 2025
Abstract
We reassess the main conclusions of Boehmer et al. (2021, BJZZ) on the predictive power of retail order imbalance (ROI) for future stock returns in the more recent 2016-2021 period and/or using the alternative quote midpoint (QMP) method. The predictive power of ROI based on the BJZZ algorithm weakens in the 2016-2021 period. Specifically, past ROI no longer predicts weekly returns on large-cap stocks, and the long-short strategy is unprofitable. Interestingly, using the QMP method not only upholds BJZZ’s main conclusions in their original sample but also offers stronger support for these conclusions during the 2016–2021 period compared to the BJZZ algorithm.
Recommanded citation: Ardia, D., Aymard, C. and Cenesizoglu, T. (2025) Revisiting Boehmer et al. (2021): Recent Period, Alternative Method, Different Conclusions, Financial Markets and Portfolio Management (forthcoming)