What Can Analysts Learn from Artificial Intelligence about Fundamental Analysis?

62 Pages Posted: 8 Jan 2021 Last revised: 8 Feb 2023

See all articles by Oliver Binz

Oliver Binz

ESMT European School of Management and Technology

Katherine Schipper

Duke University - Fuqua School of Business

Kevin Standridge

Duke University, Fuqua School of Business, Students

Date Written: January 20, 2022

Abstract

We apply machine learning to accommodate the nonlinearities that prevented Nissim and Penman (NP, 2001) from estimating their framework. We obtain more accurate out-of-sample profitability forecasts than those derived from a random walk and linear estimation and find that investing strategies based on intrinsic values generated from those profitability forecasts yield risk-adjusted returns. Our holistic estimation approach allows us to jointly analyze five financial statement analysis design choices discussed but not analyzed in NP to provide insights for the teaching and practice of fundamental analysis. First, we confirm that focusing on core items improves forecast accuracy. Second, and in contrast to prior research that examines certain profitability disaggregations individually, we find mixed evidence that increasingly granular disaggregation incrementally improves performance, holding other financial statement analysis design choices constant. Third, we provide new evidence that using a long series of historical information impairs forecast accuracy. Fourth and fifth, we find risk-adjusted returns are greatest when forecasting operating items and using a forecast horizon of five years. The benefits of greater model complexity and nonlinear estimation are pronounced for firms with extreme profitability levels and during the beginning and the end of firms’ lifecycles.

Keywords: Financial Statement Analysis, Machine Learning, Earnings Forecasting

JEL Classification: C53, G10, M41

Suggested Citation

Binz, Oliver and Schipper, Katherine and Standridge, Kevin, What Can Analysts Learn from Artificial Intelligence about Fundamental Analysis? (January 20, 2022). Available at SSRN: https://ssrn.com/abstract=3745078 or http://dx.doi.org/10.2139/ssrn.3745078

Oliver Binz (Contact Author)

ESMT European School of Management and Technology ( email )

Schlossplatz 1
10117 Berlin
Germany

Katherine Schipper

Duke University - Fuqua School of Business ( email )

Box 90120
Durham, NC 27708-0120
United States

Kevin Standridge

Duke University, Fuqua School of Business, Students ( email )

Durham, NC
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
1,219
Abstract Views
4,103
Rank
31,198
PlumX Metrics