Oz Shy
Working Paper 2020-8a
June 2020- (Revised November 2022)

Full text Adobe PDF file format

Abstract: The study of consumer payment choice at the point of sale involves classifications of payment methods such as cash, credit cards, debit cards, prepaid cards, checks, and electronic payments withdrawn from consumers’ bank account. I compare alternative methods for studying consumer payment choice using some machine learning techniques applied to the 2021 survey and diary data on consumer payment choice. The results are then compared to the more traditional logistic regression methods. Machine learning techniques have advantages in generating predictions of payment choice, in visualization of the results, and in application to high-dimensional data. The logistic regression approach has an advantage in interpreting the probability that a buyer uses a specific payment instrument.

JEL classification: C19, E42

Key words: studying consumer payment choice, point of sale, statistical learning, machine learning


The author thanks participants at the Bank of Canada Retail Payments Workshop held in Ottawa October 24–26, 2018, for valuable comments on an earlier draft. The views expressed here are those of the author and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System. Any remaining errors are the author’s responsibility.

Please address questions regarding content to Oz Shy, Research Department, Federal Reserve Bank of Atlanta, 1000 Peachtree St. NE, Atlanta, GA 30309.

Federal Reserve Bank of Atlanta working papers, including revised versions, are available on the Atlanta Fed's website at www.frbatlanta.org. Click "Publications" and then "Working Papers." To receive e-mail notifications about new papers, use frbatlanta.org/forms/subscribe.