Machine Learning

AgPa #72: Machine-Reading of Private Equity Prospectuses

Limited Partners versus Unlimited Machines: Artificial Intelligence and the Performance of Private Equity Funds (2023)
Reiner Braun, Borja Fernández Tamayo, Florencio López-de-Silanes, Ludovic Phalippou, Natalia Sigrist
CEFS Research Paper, URL/SSRN

This week’s AGNOSTIC Paper is somewhat outside my major area of competence, but I think it is a good example where we are heading to in the investment industry. Over the last years, it became quite standard that investors use the latest tools of machine learning to analyze non-quantitative information like text or images at a scale that hasn’t been possible before. So far, however, the efforts were mostly focused on public markets. In their not yet published working paper, this week’s authors show that there seems to be also a lot of potential for such methods in private markets.

  • Portfolio Company, Management Team, Investment Opportunity – The most common words of PE-managers
  • The complexity of PE-fund documents is related to fundraising success and performance
  • Machine learning and text data helps to select PE-funds
  • The machines seem to pick up meaningful concepts

Read the Full Post

AgPa #68: Machine-Learned Manager Selection (4/4)

A Cross-Sectional Machine Learning Approach for Hedge Fund Return Prediction and Selection (2021)
Wenbo Wu, Jiaqi Chen, Zhibin (Ben) Yang, Michael L. Tindall
Management Science 67(7), URL/SSRN

The fourth and at least for the moment final AGNOSTIC Paper on Machine Learned Manager Selection. After examining equity mutual funds in the last three papers, this week‘s authors provide an interesting out-of-sample test and explore machine learning models for selecting hedge funds. Importantly, this week‘s paper appeared in one of the leading business journals already back in 2021. This increases the likelihood that the results are actually robust and strengthens the evidence.

  • Machine learning helps to identify outperforming hedge funds
  • Risk measures and VIX-correlations are the most important features

Read the Full Post

AgPa #67: Machine-Learned Manager Selection (3/4)

Selecting Mutual Funds from the Stocks They Hold: A Machine Learning Approach (2020)
Bin Li, Alberto G. Rossi
SSRN Working Paper, URL

The third AGNOSTIC Paper on the application of machine learning in manager selection. This week’s paper is very similar to AgPa #65 and AgPa #66, and again examines the data on US mutual funds. Despite somewhat different methodology, the results point in a similar direction. This, of course, increases the evidence that machine learning is actually useful for manager selection…

  • Machine learning helps to identify outperforming funds
  • The best and worst funds share common characteristics
  • Trading Frictions and Momentum are the most relevant variables

Read the Full Post

AgPa #66: Machine-Learned Manager Selection (2/4)

Machine Learning and Fund Characteristics Help to Select Mutual Funds with Positive Alpha (2023)
Victor DeMiguel, Javier Gil-Bazo, Francisco J. Nogales, Andre A. P. Santos
SSRN Working Paper, URL

The second AGNOSTIC Paper on the application of machine learning in manager selection. This week’s paper follows essentially the same idea as Kaniel et al. (2022) in AgPa #65. The authors also examine a comprehensive sample of US mutual funds and although they use slightly different methodology, arrive at generally similar conclusions. This, of course, increases the evidence that machine learning is indeed helpful for manager selection…

  • Machine learning helps to identify outperforming funds
  • Past performance and measures of activeness are the most relevant variables
  • Given their alpha, machine-selected funds remain too small

Read the Full Post

AgPa #65: Machine-Learned Manager Selection (1/4)

Machine-Learning the Skill of Mutual Fund Managers (2022)
Ron Kaniel, Zihan Lin, Markus Pelger, Stijn Van Nieuwerburgh
NBER Working Paper 29723, URL

To conclude the posts on manager selection, at least for the moment, I will dive into one of the most recent research frontiers in this area. Since the application of machine learning in investment management has been intensively studied among equities for more than three years now, it is not surprising that researchers also start to apply such algorithms to other asset classes. A natural candidate for this are equity mutual funds and this is exactly where this and the next four week’s AGNOSTIC Papers come in.

  • Machine learning helps to identify outperforming funds
  • Less is more – not all information is necessary
  • Alpha is easier to predict than total returns

Read the Full Post

AgPa #55: Backtests in the Age of Machine Learning

A Backtesting Protocol in the Era of Machine Learning (2019)
Rob Arnott, Campbell R. Harvey, Harry Markowitz
The Journal of Financial Data Science Winter 2019, URL/SSRN/PDF

I have already written about the pitfalls of research in asset management and the importance of good research practices for the application of machine learning. This week’s AGNOSTIC Paper takes this idea even further and provides a seven-point protocol for empirical research in finance.

Exhibit 2 of Arnott et al. (2019).

Read the Full Post

AgPa #49: Machine Learning in Quant Asset Management

How Can Machine Learning Advance Quantitative Asset Management? (2023)
David Blitz, Tobias Hoogteijling, Harald Lohre, Philip Messow
The Journal of Portfolio Management Quantitative Tools 2023, URL/SSRN

This week’s AGNOSTIC Paper is a broad overview about machine learning in investment management. The authors outline the benefits and pitfalls of machine learning compared to “traditional” econometrics and present several use cases in the world of (quantitative) asset management. They also provide ideas for research governance to keep those powerful methods under control.

  • Benefits and pitfalls of machine learning in finance
  • Use cases of machine learning in asset management
  • Keeping it under control: research governance and protocol

Read the Full Post

AgPa #48: Investable Machine Learning for Equities

Investable and Interpretable Machine Learning for Equities (2022)
Yimou Li, Zachary Simon, David Turkington
The Journal of Financial Data Science Winter 2022, 4(1), URL

Regular readers of this blog know that machine learning in asset management is one of my favorite topics and I recently found new interesting material. This week’s AGNOSTIC Paper is the first of two studies and examines an important issue with machine learning models in great detail: interpretability…

  • Machine learning models outperform simpler methods
  • Different models learn different investment approaches

Read the Full Post

AgPa #41: US Factors before 1926

The Cross-Section of Stock Returns before CRSP (2023)
Guido Baltussen, Bart van Vliet, Pim van Vliet
SSRN Working Paper, URL

This week’s AGNOSTIC Paper is an unprecedented out-of-sample test of the four major factors (Momentum, Value, Low-Risk, Size). The authors construct a novel dataset of US stocks that reaches from 1866 to 1926. It therefore extends the extensively studied CRSP dataset by 60 years.

  • Momentum, Value, and Low-Risk were there before 1926
  • Factors weren’t stronger before 1926
  • Machine learning models find the same factors

Read the Full Post

AgPa #32: Agnostic Fundamental Analysis (3/3)

Boosting agnostic fundamental analysis: Using machine learning to identify mispricing in European stock markets (2022)
Matthias X.Hanauer, Marina Kononova, Marc Steffen Rapp
Finance Research Letters 48, URL/SSRN

The third and final post about agnostic fundamental analysis. This week’s AGNOSTIC Paper challenges the simple linear methodology and introduces vastly improved valuation models…

  • More sophisticated valuation models yielded better performance
  • Different models emphasize different fundamental variables

Read the Full Post