Challenges and opportunities in high-dimensional choice data analyses
- PDF / 180,067 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 5 Downloads / 210 Views
Challenges and opportunities in high-dimensional choice data analyses Prasad Naik & Michel Wedel & Lynd Bacon & Anand Bodapati & Eric Bradlow & Wagner Kamakura & Jeffrey Kreulen & Peter Lenk & David M. Madigan & Alan Montgomery
Published online: 31 May 2008 # Springer Science + Business Media, LLC 2008
Abstract Modern businesses routinely capture data on millions of observations across subjects, brand SKUs, time periods, predictor variables, and store locations, thereby generating massive high-dimensional datasets. For example, Netflix has Prasad Naik and Michel Wedel are co-chairs. P. Naik University of California, Davis, Davis, CA, USA e-mail: [email protected] M. Wedel (*) University of Maryland, College Park, MD, USA e-mail: [email protected] L. Bacon Polimetrix Inc., Palo Alto, CA, USA A. Bodapati University of California, Los Angeles, Los Angeles, CA, USA E. Bradlow University of Pennsylvania, Philadelphia, PA, USA W. Kamakura Duke University, Durham, NC, USA J. Kreulen IBM Almaden Research Center, San Jose, CA, USA P. Lenk University of Michigan, Ann Arbor, MI, USA D. M. Madigan Columbia University, New York, NY, USA A. Montgomery Carnegie Mellon University, Pittsburgh, PA, USA
202
Market Lett (2008) 19:201–213
choice data on billions of movies selected, user ratings, and geodemographic characteristics. Similar datasets emerge in retailing with potential use of RFIDs, online auctions (e.g., eBay), social networking sites (e.g., mySpace), product reviews (e.g., ePinion), customer relationship marketing, internet commerce, and mobile marketing. We envision massive databases as four-way VAST matrix arrays of Variables×Alternatives×Subjects×Time where at least one dimension is very large. Predictive choice modeling of such massive databases poses novel computational and modeling issues, and the negligence of academic research to address them will result in a disconnect from the marketing practice and an impoverishment of marketing theory. To address these issues, we discuss and identify the challenges and opportunities for both practicing and academic marketers. Thus, we offer an impetus for advancing research in this nascent area and fostering collaboration across scientific disciplines to improve the practice of marketing in information-rich environment. Keywords Challenges . Opportunities . High-dimensional choice data analyses . Modern businesses . Four-way VAST matrix arrays
1 Introduction Advances in computing power have revolutionized marketing practice. Companies not only possess massive database on billions of consumer choices, user ratings, and geodemographics, but also recognize that the predictive modeling of massive choice data offers potentially high return on investment. As firms become customerfocused, they amass customer databases by recording various aspects resulting from interactions with customers. A fundamental consequence of building massive databases to marketing is the shift in focus from reaching anonymous consumers (e.g., in mass marketing via television) to targeting identi
Data Loading...
