Homepage of Eric M. Schwartz
Last updated 2021-06-05
Curriculum Vita.
Welcome to Eric Schwartz's simple website!
Eric Schwartz is an Associate Professor of Marketing, with tenure, at the Stephen M. Ross School of Business at the University of Michigan. He is a data scientist applying research in statistics, machine learning, and econometrics to a range of problems. These span problems in customer analytics for marketing, such as A/B testing methods, native advertising, streaming media, and valuing customers, as well as in optimal resource allocation for public health. In the classroom, Professor Schwartz focuses on the quantitative aspects of marketing, including an elective on customer lifetime value as well as the introductory core marketing course. He is also a co-founder of BlueConduit, a social venture spun out of the University of Michigan applying machine learning research developed during the Flint Water Crisis to find lead pipes to cities and utilities across North America. For more biographical information, see below.
What's New
- [2021 September] Talk at (remote) Marketing Modelers seminar series
- [2021 August] Interview with Crain's Detroit.
- [2021 August] Submitted revision of "Finding the Sweet Spot" paper to Journal of Marketing Research (with Prashant Rajaram and Puneet Manchanda).
- [2021 July] Submitted new paper "A/B Testing Deception" to Marketing Science (with Michael Braun).
- [2021 June] Talk at (remote) Marketing Science Conference
- [2021 February] Talk at (remote) University of Texas, Dallas, Bass FORMS Conference
- [2021 January] Seminar talk at (remote) University of Colorado, Boulder, Leeds Marketing
- [2021 February] Interview on NPR's Science Friday radio/podcast show
- [2021 January] Featured in WIRED magazine article
- Aribarg, Anocha and Eric M. Schwartz (2020). Native advertising in online news: Tradeoffs among clicks, brand recognition and website trustworthiness, *Journal of Marketing Research*, 57(1), 20-24.
Journal Link.
PDF.
BibTeX.
- Finalist for Paul E. Green Award (2020) for best paper in Journal of Marketing Research
- Proserpio, D., Hauser, J. R., Liu, X., Amano, T., Burnap, A., Guo, T., Lee, D., Lewis, R. A., Misra, K., Schwartz, E. M., Timoshenko, A., Xu, L., and Yoganarasimhan, H., (2020) Soul and Machine (Learning), Marketing Letters, 31(4), Special Issue for 11th Triennial Invitational Choice Symposium, 393-404. Journal Link. PDF.
- Misra, Kanishka, Eric M. Schwartz, Jacob D. Abernethy (2019). Dynamic online pricing with incomplete information using multi-armed bandit experiments. Marketing Science, 38(2), 226-252.
Journal Link.
PDF.
BibTeX.
Google Scholar.
- Finalist for the John D. C. Little Award (2019) for the best marketing paper in Marketing Science, Management Science, and all INFORMS journals.
- Schwartz, Eric M., Bradlow, Eric T., and Fader, Peter S. (2017). Customer acquisition via display advertising using multi-armed bandit experiments. Marketing Science, 36(4), 500-522.
Journal Link.
PDF.
BibTeX.
Google Scholar.
- Schwartz, Eric M., Bradlow, Eric T., and Fader, Peter S. (2014). Model selection using database characteristics: Developing a classification tree for longitudinal incidence data. Marketing Science, 33(2), 188-205.
Journal Link.
PDF.
BibTeX.
Google Scholar.
Press Release.
- Berger, Jonah, and Eric M. Schwartz (2011). What drives immediate and ongoing word of mouth? Journal of Marketing Research, 48 (5), 869-880.
Journal Link.
PDF.
BibTeX.
Google Scholar. Featured in Contagious .
- Braun, Michael, and Eric M. Schwartz (2021). A/B Test Deception: Divergent Delivery, Ad Response Heterogeneity, and Erroneous Inferences in Online Advertising Field Experiments
.
- Rajaram, Prashant, Puneet Manchanda, and Eric M. Schwartz (2021). Finding the Sweet Spot: Ad Targeting on Streaming Media.
- Submitted revision to Journal of Marketing Research
- PDF on SSRN.
- Presented at AI and Marketing Science Workshop at AAAI Conference 2018 (2018-Feb-02)
- Schwartz, Eric M., Jacob D. Abernethy, Jared Webb (2019). Active Learning for Sequential Household-level Targeted Intervention: An Application to Find Lead Pipes in Flint, Michigan.
- Schwartz, Eric M., Kenneth Fairchild, Bryan Orme, Alexander Zaitzeff (2019). Active Learning for Ranking and Selection: Bandit MaxDiff for Idea Screening.
-
Abernethy, Jacob D., Alex Chojacki^, Arya Farahi^, Eric M. Schwartz, Jared Webb^* (2018). ActiveRemediation: The Search for Lead Pipes in Flint, Michigan.
KDD 2018, Proceedings of SIGKDD Conference on Knowledge Discovery and Data Mining, London, England, U.K. *Alphabetical order. ^Student.
-
Chojnacki, Alex^, Chengyu Dai^, Arya Farahi^, Guangsha Shi^, Jared Webb^, Daniel T. Zhang^, Jacob Abernethy, Eric M. Schwartz* (2017). A Data Science Approach to Understanding Residential Water Contamination in Flint. KDD 2017, Proceedings of SIGKDD Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada. ^Student. *Students first, then faculty; alphabetical order.
- PDF on Arxiv
- Work involving students of the Michigan Data Science Team (MDST).
- Press about Flint's lead levels in its water:
- Google funds research and app development, joint with U-M Flint Computer Science and Michigan Data Science Team (2016-May-3) reported in: Chicago Tribune, Tech Crunch, Gizmodo, The Hill, Detroit Free Press, MLive, Michigan Radio, The University Record, Michigan Engineering News.
- Jacob Abernethy, Cyrus Anderson, Chengyu Dai, Arya Farahi, Linh Nguyen, Adam Rauh, Eric M. Schwartz, Wenbo Shen, Guangsha Shi, Jonathan Stroud, Xinyu Tan, Jared Webb, Sheng Yang* (2016). Flint Water Crisis: Data-Driven Risk Assessment Via Residential Water Testing, in Proceedings of 2nd Annual Bloomberg Conference Data for Good Exchange (D4GX 2016), NY, NY.
*alphabetical order.
- Jake Abernethy, Cyrus Anderson, Alex Chojnacki, Chengyu Dai, John Dryden, Eric M. Schwartz, Wenbo Shen, Jonathan Stroud, Laura Wendlandt, Sheng Yang, Daniel Zhang* (2016). Data Science in Service of Performing Arts: Applying Machine Learning to Predicting Audience Preferences, in Proceedings of 2nd Annual Bloomberg Conference Data for Good Exchange (D4GX 2016), NY, NY.
*alphabetical order.
- Fairchild, Kenneth, Bryan Orme, Eric M. Schwartz (2015), Bandit Adaptive MaxDiff Designs for Huge Number of Items, in Proceedings of 2015 Sawtooth Software Conference, 105-117.
- PDF. Work in Collaboration with Sawtooth Software.
Current courses
- Marketing Management, MKT 503, MBA Core (2021 Fall)
- Living Business Leadership Experience, MBA (2018-present)
Past courses
- Marketing Management, MKT 503, MBA Core (2017 Fall-present)
- Marketing Management, MKT 300, BBA Core (2013-2016 Fall)
- Customer Analytics: Measuring and Managing Customer Value, MKT 626, Grad elective (2021-present)
- Customer Analytics: Measuring and Managing Customer Value, MKT 426, Undergrad elective (2020-present)
- Living Business Leadership Experience, MBA and BBA (2018-present)
Teaching interests
- Customer-base analysis and customer lifetime value; data science, model building, and statistical machine learning for customer analytics; marketing research and experimental design in marketing practice; action-based learning.
Teaching materials developed
- Data science work to find Flint's lead service lines:
- 2019:
NPR OnPoint (2019-08-22),
The Atlantic (2019-01-02),
American Civil Liberties Union (video: 2019-04-10),
ACLU / Flint Journal (2019-04-12),
Natural Resources Defense Council (2019-02-12),
Irish Times (2019-05-02), Michigan Radio
(2019-04-12,
audio: 2019-04-25),
Weather.com (2019-04-29),
Flint Journal
(2019-02-12,
2019-04-11),
Detroit News
(2019-02-12),
Now This News (video: 2019-04-28)
- 2018:
U.S. District Court filing (Concerned Pastors et al. v Kohuri et al., 2018-10-01),
Flint City Council Meeting (2018-12-05),
Bridge Michigan (2018-09-04),
New Scientist (2018-08-22),
Flint Journal (
2018-10-18
,
2018-11-26
),
Bloomberg: Environment (2018-09-21),
Bloomberg: Law and Business (2018-08-09),
Michigan Today (2018-08-20, Video)
- 2016-17: Co-authored report with City of Flint and FAST Start team for service line replacement,
City of Flint Press Release (2016-Dec-01),
CBS Local (2016-Dec-02),
MLive (2016-Dec-01),
The Detroit News (2016-Dec-01,
AccuWeather (2017-May-04),
New York Times (2017-Mar-27),
Wikipedia Citation (Accessed 2018-Jan-01)
- Data science work to predict levels of lead in drinking water in Flint:
- Co-authored op-ed (with Jacob Abernethy) How big data and algorithms are slashing the cost of fixing Flint's water crisis. The Conversation (2016-Sep-08) republished/reported in --
Scientific American ,
Business Insider (2016-Sep-08),
Associated Press (2016-Sep-08),
USA Today (2016-Sep-08),
Government and Technology (date),
Detroit Free Press (
2016-Sep-08,
2016-Sep-09,
2016-Sep-26, and
2016-Sep-11),
GreedBiz (2016-Sep-08),
Civics Analytics and Urban Intelligence on Medium (2016-Oct-30),
U-M Office of Government Relations - Michigan Impact (2016-Oct-30)
Eric Schwartz is an Associate Professor of Marketing (with tenure) at the Stephen M. Ross School of Business at the University of Michigan. Professor Schwartz's expertise focuses on predicting customer behavior, understanding its drivers, and examining how firms actively acquire customers and manage their relationships through interactive marketing experiments and adaptive data collection. His current projects aim to optimize firms' A/B testing and adaptive marketing experiments using a multi-armed bandit framework, often working with companies and organizations. His broader research in customer analytics stretches across managerial applications, including online experiments, online advertising, dynamic pricing, native advertising, streaming video binge viewing, and word-of-mouth. The quantitative methods he uses are primarily machine learning, active learning, Bayesian statistics, and field experiments. Applying those same methods elsewhere, he also works on public policy problems focused on health and safety. His work has been recognized with awards, including ISMS John D. C. Little Best Paper, ISMS Doctoral Dissertation Proposal Competition Winner, and KDD Applied Data Science Best Student Paper. He is a member of the Editorial Review Board of INFORMS journal, Marketing Science. At Ross, he was the Arnold M. and Linda T. Jacob Faculty Fellow 2018-19. Before joining the Michigan Ross faculty in 2013, Professor Schwartz earned his Ph.D. in Marketing from the Wharton School and a B.A. in Mathematics and Hispanic Studies, all from the University of Pennsylvania.
BibTeX Citations
@article{aribargschwartz2020native,
title={Native Advertising in Online News: Trade-Offs Among Clicks, Brand Recognition, and Website Trustworthiness},
author={Aribarg, Anocha and Schwartz, Eric M},
journal={Journal of Marketing Research},
volume={57},
number={1},
pages={20--34},
year={2020}
}
@article{msa2018banditpricing,
title={Customer acquisition via display advertising using
multi-armed bandit experiments},
author={Misra, Kanishka and Schwartz, Eric and Jacob D. Abernethy},
journal={Marketing Science},
volume={Forthcoming},
year={2018},
publisher={INFORMS}
}
@article{schwartzetal2017bandit,
title={Customer acquisition via display advertising using
multi-armed bandit experiments},
author={Schwartz, Eric M and Bradlow, Eric T and Fader, Peter S},
journal={Marketing Science},
volume={36},
number={4},
pages={500--522},
year={2017},
publisher={INFORMS}
}
@article{schwartzetal2014hmmrf,
title={Model selection using database characteristics: Developing a classification tree for longitudinal incidence data},
author={Schwartz, Eric M and Bradlow, Eric T and Fader, Peter S},
journal={Marketing Science},
volume={33},
number={2},
pages={188--205},
year={2014},
publisher={INFORMS}
}
@article{bergerschwartz2011wom,
title={What drives immediate and ongoing word of mouth?},
author={Berger, Jonah and Schwartz, Eric M},
journal={Journal of Marketing Research},
volume={48},
number={5},
pages={869--880},
year={2011},
publisher={American Marketing Association}
}
(End)
This page was typed by hand and written in HyperText Markup Language (HTML). That means Web 1.0, 1990s style, without any fancy apps and slick Web 2.0 style graphics. No WhatYouSeeIsWhatYouGet editors are needed. This is a flat website in a single page. You can see all contents with full transparency with Show Page Source / Inspect Element in your browser. I used the template provided by Frank da Cruz.
~ Eric Schwartz ~