Redefining Risk-Adjusted Returns of PE Funds through Advanced Analytics

Advisor: Dr. Sara Boni & Dr. Max Knicker

Type: Bachelor’s or Master’s Thesis
Start: Sep. 2025

 

Overview

Private Equity (PE) has emerged as one of the fastest-growing segments of the global asset management industry, with institutional and individual investors increasingly allocating capital to this asset class. Yet, measuring the performance of PE funds remains a central challenge. Unlike public market securities, PE investments are illiquid, opaque, and characterized by irregular cash flows, making standard risk-return measures such as the Sharpe ratio inadequate.

Previous literature has proposed several approaches to estimate and adjust returns in the presence of illiquidity and time-varying risk exposures. For example, valuation methodologies such as the Public Market Equivalent (PME), Internal Rate of Return (IRR), and Direct Alpha are widely used but often criticized for their sensitivity to assumptions and lack of robustness across fund vintages and market cycles.

Recent advances in data analytics and computational methods provide an opportunity to revisit these issues. Machine learning algorithms, Bayesian updating, and advanced time-series econometrics (including nowcasting and strip-based valuation methods) can yield more dynamic and accurate measures of risk-adjusted returns in PE funds.

This thesis aims to bridge traditional approaches with modern empirical techniques, providing a systematic review of the literature and implementing alternative measures for assessing risk-adjusted performance in Private Equity. The final contribution will consist of both a survey of existing methodologies and a hands-on empirical application using Python and Matlab to compute new return measures on real or simulated data.

Research Objectives

1. Literature Review
   • Summarize the classical methods for measuring PE returns (IRR, PME, Direct Alpha, Kaplan-Schoar measures).
   • Review recent contributions that adapt public market tools to PE settings (e.g., strip valuation, nowcasting NAVs).
   • Highlight the shortcomings of existing approaches, especially with respect to risk adjustment.
2. Methodological Innovation
   • Explore how advanced analytics (machine learning, robust optimization, quantile regressions, Bayesian methods, etc.) can refine risk-adjusted return measures.
   • Develop and implement at least one new metric for PE performance evaluation.
3. Empirical Application
   • Use available datasets (or simulated PE cash flow data if confidentiality constraints apply).
   • Compare traditional and newly developed risk-adjusted performance measures.
   • Assess robustness across different fund vintages, strategies (buyouts, venture capital, growth equity), and macroeconomic conditions.

Requirements

• Strong interest in quantitative finance and private markets.
• Motivation for empirical research and curiosity about the mechanics of PE funds.
• Solid programming skills in Python or Matlab.
• Attention to detail, independence, and willingness to work with complex datasets.

Literature

• Gupta, A. and Van Nieuwerburgh, S. (2021). Valuing Private Equity Investments Strip by Strip. The Journal of Finance, 76: 3255–3307. [DOI: 10.1111/jofi.13073]
• Brown, G. W., Ghysels, E., Gredil, O. R. (2023). Nowcasting Net Asset Values: The Case of Private Equity. The Review of Financial Studies, 36(3): 945–986. [DOI: 10.1093/rfs/hhac045]
• Harris, R. S., Jenkinson, T., & Kaplan, S. N. (2014). Private Equity Performance: What Do We Know? Journal of Finance, 69(5): 1851–1882.

Contact & Application

If you are interested in writing your thesis on this topic, please include this project in your application and briefly outline your motivation, research interests, and programming experience. Please note that the thesis topic can be tailored to your interests—for example, by focusing more on econometric modeling, machine learning, or empirical applications.