Forecasting Earnings Using k-Nearest Neighbors

48 Pages Posted: 19 Feb 2021 Last revised: 1 Dec 2023

See all articles by Peter D. Easton

Peter D. Easton

University of Notre Dame - Department of Accountancy

Martin M. Kapons

University of Amsterdam - Amsterdam Business School

Steven J. Monahan

INSEAD; University of Utah

Harm H. Schütt

Tilburg University - Tilburg School of Economics and Management

Eric H. Weisbrod

University of Kansas - School of Business

Date Written: July 26, 2021

Abstract

We use a simple k-nearest-neighbors algorithm (hereafter, k-NN*) to forecast earnings. k-NN* forecasts of one-, two-, and three-year ahead earnings are more accurate than those generated by popular extant forecasting approaches. k-NN* forecasts of two- and three-year (one-year) ahead EPS and aggregate three-year EPS are more (less) accurate than those generated by analysts. The association between the unexpected earnings implied by k-NN* and the contemporaneous market-adjusted return (i.e., the earnings association coefficient, EAC) is positive and exceeds the EAC on unexpected earnings implied by alternate approaches. A trading strategy that is long (short) firms for which k-NN* predicts positive (negative) earnings growth earns positive risk-adjusted returns that exceed those earned by similar trading strategies that are based on alternate forecasts. The k-NN* algorithm generates an empirically reliable ex ante indicator of forecast accuracy that identifies situations when the k-NN* EAC is larger and the k-NN* trading strategy is more profitable.

Keywords: earnings, forecasting, machine learning

JEL Classification: C21, C53, G17, M41

Suggested Citation

Easton, Peter D. and Kapons, Martin M. and Monahan, Steven J. and Monahan, Steven J. and Schütt, Harm H. and Weisbrod, Eric H., Forecasting Earnings Using k-Nearest Neighbors (July 26, 2021). Available at SSRN: https://ssrn.com/abstract=3752238 or http://dx.doi.org/10.2139/ssrn.3752238

Peter D. Easton (Contact Author)

University of Notre Dame - Department of Accountancy ( email )

Mendoza College of Business
Notre Dame, IN 46556-5646
United States
574-631-6096 (Phone)
574-631-5127 (Fax)

Martin M. Kapons

University of Amsterdam - Amsterdam Business School ( email )

Spui 21
Amsterdam, 1018 WB
Netherlands

Steven J. Monahan

INSEAD ( email )

Boulevard de Constance
PMLS 1.24
F-7705 Fontainebleau Cedex, 77305
France
+33 1 60 72 92 14 (Phone)
+33 1 60 72 92 53 (Fax)

HOME PAGE: http://www.insead.edu/facultyresearch/faculty/profiles/smonahan/

University of Utah ( email )

1645 E Campus Center Dr
Salt Lake City, UT 84112-9303
United States

Harm H. Schütt

Tilburg University - Tilburg School of Economics and Management ( email )

PO Box 90153
Tilburg, 5000 LE Ti
Netherlands

Eric H. Weisbrod

University of Kansas - School of Business ( email )

1300 Sunnyside Avenue
Lawrence, KS 66045
United States

HOME PAGE: http://https://business.ku.edu/eric-weisbrod

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

Paper statistics

Downloads
456
Abstract Views
2,728
PlumX Metrics