Commentary—A Logit Model of Brand Choice Calibrated on Scanner Data: A 25th Anniversary Perspective
Peter M. Guadagni,
John D. C. Little
Bonanza Steet Books, Walnut Creek, California 94596
Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts 02142
guadagni{at}alum.mit.edu
jlittle{at}mit.edu
Guadagni and Little (1983) had a surprising (to the authors) number of citations, presumably because it was the first paper to build a useful model with UPC scanner panel data. More surprising (but not to the authors) was that the model, compared to those in most academic papers, found widespread practical application. The reason for this was that a small, entrepreneurial consulting firm developed and sold applications based on the model. The paper also generated a substantial follow-on of academic literature. Examples illustrate a few of the directions in which later research went.
Key Words: brand choice; multinomial logit; UPC data; marketing science practice
History: Received: November 6, 2007;
Copyright © 2008 by INFORMS.