TY - JOUR T1 - Fine-Tuning Private Equity Replication Using Textual Analysis JF - The Journal of Financial Data Science SP - 111 LP - 121 DO - 10.3905/jfds.2019.1.1.111 VL - 1 IS - 1 AU - Ananth Madhavan AU - Aleksander Sobczyk Y1 - 2019/01/31 UR - https://pm-research.com/content/1/1/111.abstract N2 - In this article, the authors use textual analysis to create an investable, dynamic portfolio to mimic the factor characteristics of private equity. First, using textual analysis, they identify firms taken private by those firms in the 10-year period ending June 2018. Second, they use a multifactor model to measure the cross-sectional factor exposures of firms immediately prior to the announcement that they were being acquired by a private equity firm. Finally, they use holdings-based optimization to build a liquid, investible, long-only portfolio that dynamically mimics the factor characteristics of the portfolio of stocks that were taken private. Practitioner applications include interim beta solutions for investors (including venture capital and private equity firms) seeking to deploy excess cash, mitigate underfunding risk, and manage capital calls.TOPICS: Big data/machine learning, factors, risk premia, private equity ER -