Predicting Revenues With the Multiplier Heuristic

21 Pages Posted: 26 Mar 2020

See all articles by Florian M. Artinger

Florian M. Artinger

Max Planck Institute for Human Development

Nikita Kozodoi

Amazon Web Services, Inc.

Julian Runge

Duke University (Visiting Scholar)

Date Written: February 28, 2020

Abstract

Using machine learning to forecast revenues per customer, product, or store has become a major industry. It contrasts with a simple, intuitive managerial heuristic: multiply the revenue observed in the first t days by a constant. We test the predictive accuracy of this multiplier heuristic in 20 data sets. Surprisingly, the heuristic performs overall on par with machine learning models. We identify three central drivers when the heuristic can even outperform these models: limited sample size, time-to-prediction, and changes across time. The results provide insights when to rely on heuristics and managerial intuition or when on machine learning.

Suggested Citation

Artinger, Florian M. and Kozodoi, Nikita and Runge, Julian, Predicting Revenues With the Multiplier Heuristic (February 28, 2020). Available at SSRN: https://ssrn.com/abstract=3546017 or http://dx.doi.org/10.2139/ssrn.3546017

Florian M. Artinger (Contact Author)

Max Planck Institute for Human Development ( email )

Lentzeallee 94
D-14195 Berlin, 14195
Germany

Nikita Kozodoi

Amazon Web Services, Inc.

Julian Runge

Duke University (Visiting Scholar) ( email )

Box 90120
Durham, NC 27708-0120
United States

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

Paper statistics

Downloads
109
Abstract Views
753
Rank
450,735
PlumX Metrics