2018 Advanced Research Techniques (ART) Forum

Ohio State University
Columbus, OH 43210
6/21/2018 8:00 AM - 6/22/2018 5:00 PM
Register before 5/24/2018 11:59 PM for early registration fee

Data scientists, marketing professionals and academics attending the Advanced Research Techniques (ART) Forum find a network of colleagues who are serious about developing tools that can solve the next generation of problems in the field of marketing. 

New hands-on sessions, focusing on repeatable research, will make sure you leave with practical takeaways--and the right code--for applying sophisticated methodologies and quantitative techniques to support strategic marketing decisions.

At the ART Forum, you’ll find a community not only dedicated to advancing their research skills, but also ready to help support each other through the process. This in an intimate event where you’ll have access to speakers and other smart minds throughout the entire event.

2018 Schedule At A Glance​

2018 Program Committee

Greg Allenby - Chair

Helen C. Kurtz Chair in Marketing, The Ohio State University

Jeff Dotson
Associate Professor of Marketing, Brigham Young University

Elea Feit
Assistant Professor of Marketing, Drexel University


Who Attends This Conference

Since its inception in 1990, the conference has focused on how sophisticated methodologies and quantitative techniques can support strategic and tactical marketing decisions. The attendees are experienced research practitioners who use advanced methods, and marketing academics interested in industry problems.

AMA Cancellation Policies​

Wednesday, June 20 – Tutorials 

Advanced Tutorials

12:30 – 2:30 Volumetric Choice Models – Nino Hardt, Ohio State University

Purchases in many product categories (e.g., yoghurt, pizza, or other packaged goods) are characterized by quantity demand and variety seeking. In contrast to discrete choice conjoint, volumetric conjoint allows respondents to state their demand for multiple products. In this tutorial, participants will be introduced to volumetric conjoint analysis. This is includes design considerations, estimation, and application to typical industry problems such as market simulation and assessing (cross-) price elasticities. We will be using the R statistical language for this tutorial, and working knowledge of R is strongly suggested. A dedicated R package for volumetric conjoint estimation and additional analysis will be supplied (pre-compiled and source).

3:00 – 5:00 A Tidy Approach to Text Analysis in R – Marc Dotson, BYU

Text analysis has become increasingly accessible in R. The recent development of the tidytext package, built on the foundation of the tidyverse, continues this trend. In this tutorial, we will cover the basics of text analysis in R consistent with the underlying philosophy of the tidyverse, including basic counts, sentiment analysis, and topic modeling. Basic fluency in R and the tidyverse is expected (e.g., see the tutorial on R and the tidyverse). Participants should have the latest version of R, RStudio, and the tidyverse installed.

Beginner Tutorials

12:30 – 2:30 Introduction to R with Tidyverse – Marc Dotson, BYU 

R is one of the most popular open-source languages for data analysis. The tidyverse is a collection of powerful packages in R for common tasks, including data cleaning and visualization. The tidyverse provides a consistent and intuitive introduction to R and serves as a foundation for a growing number of packages in R (e.g., see the tutorial on text analysis). In this tutorial, we will cover the basics of using R and the tidyverse. Before the tutorial, participants should download and install the latest version of R (cran.r-project.org) and RStudio (www.rstudio.com).

3:00 – 5:00 Introduction to HB Modeling with STAN – Elea Feit, Drexel University

Stan is a newly-developed software platform that allows users to specify and estimate a wide range of models using a powerful MCMC engine. In this tutorial, participants will be introduced to Stan and learn how to estimate a hierarchical Bayes choice model in Stan. Emphasis will be placed on good practices in Bayesian inference and hierarchical modeling.  All code for running models will be provided via GitHub. We will be accessing Stan through the R statistical language and working knowledge of R is strongly suggested. 



Thursday, June 21, 2018

7:00 – 8:20    Breakfast
8:20 – 8:30    Welcoming Introduction by Greg Allenby, Ohio State University
8:30 – 10:00    Advances in Choice Modeling I
Managerial Inference from a Direct Utility Model for Volumetric Conjoint: A Critical Look at Interview Set Up, Experimental Design, and Model Priors.

Jake Lee – Quantum Strategy, Inc

Volumetric conjoint models are relatively new. They will be critical for categories where a single forced choice doesn’t make sense. This paper will look at how various design and modeling decisions will impact managerial inference. Special attention will be paid to substitution vs category expansion.

Adaptive MaxDiff Designs for Huge Numbers of Items
Kenneth Fairchild, Zachary Anderson, and Bryan Orme – Sawtooth Software

MaxDiff (Best-Worst Scaling) has become a very popular technique in survey research for prioritizing a list of items. When the list of items grows large and when the emphasis is on prioritizing the top few items for the sample, standard level-balanced designs are wasteful and inefficient. An adaptive MaxDiff design strategy based on Thompson sampling can increase the efficiency from 2x to 3.5x, saving about 50% to 70% on data collection costs. Using either MNL or faster counting analysis, we can learn from earlier respondents to focus the efforts of later respondents on what have a high likelihood of being the more preferred items.

10:00 – 10:30    Break
10:30 – 12:00    Advances in Choice Modeling II
An Improved Latent Class (LC) Paradigm to Obtain Meaningful Segments in the Presence of Scale Confounds: Scale Adjusted Latent Class (SALC) Tree modeling

Jay Magidson – Statistical Innovations

Accounting for both mean and variance (scale) heterogeneity is important in discrete choice modeling in order to obtain estimates of preference that are free from scale confounds (strength of preference), and to obtain meaningful latent class (LC) segments that differ on these preferences. We propose a new model and related methods that extend LC tree (LCT) modeling to utilize two categorical latent variables, one to separate out the effects of scale, the other to obtain meaningful segments that differ solely in their preferences. We apply this model to MaxDiff response data containing both preference and scale heterogeneity and evaluate the effective ness of this new model to remove scale confounds and classify respondents into meaningful LC segments.

Results show that the new SALC Tree approach reveals policy relevant segments very similar to those hypothesized to exist prior to the analysis, while segments obtained with currently available methods are shown to yield less desirable results. This new modeling approach is being implemented in version 6.0 of the Latent GOLD® program.

How Content Affects Clicks: A Dynamic Model of Online Content Consumption
Inyoung Chae, David A. Schweidel, Da Young Kim – Emory University

With many consumers being exposed to news via social media platforms, news organizations are challenged to attract visitors and generate revenue during visits to their websites. For advertising-supported websites, website traffic is related directly to revenue. Using clickstream data from a major news organization, the authors develop a user-level dynamic model of clickstream behavior that takes into account the content of both headlines and stories that visitors read. 

12:00 – 1:20    Lunch
1:20 – 1:30    Introduction to Afternoon Sessions – Elea Feit, Drexel University
1:30 – 3:00    Advances in Modeling Dynamics I
Using Dynamic Linear Models (DLM) to Quantify the Time-Varying Contribution of Marketing Instruments to Sales

Ziad Elmously – Kantar

This presentation will describe the application of a Bayesian implementation of Dynamic Linear Models (DLM). Using data in the auto category, we estimate a Marketing Mix Model with time-varying covariates to capture the dynamics in the relationship between business outcomes on the one hand, and brand equity and marketing instruments on the other. We empirically demonstrate the merits of DLM, including the ability to control for extraneous factors and improve model fit statistics. Once the model is specified, the Bayesian framework enables us to quickly formulate and test variations of it via the Markov Chain Monte Carlo algorithm using commercial or freely available software. Building on the Bayesian implementation, we discuss potential extensions to model, including a multivariate version of DLM and the application to spatio-temporal models with a spatial smoothing component. We further describe the application of DLM to the estimation of Sales-Based Brand Equity (SBBE) and maximizing return on investment.

Rapid Optimization Application Development using Excel and Solver
Michael Mina – VP of Analytics and Portfolio Management at PNC Financial Services

While a number of marketing optimization tools are commercially available, many of them require changes to data and computational infrastructure that are labor-intensive and cost-prohibitive.  This presentation will discuss how segmentation can be used to reduce the complexity of large optimization problems, and how to quickly develop a simple but effective optimization application using Excel combined with Excel Solver.

Simulated data from a large catalog marketing campaign will be used and optimized for different conditions using a specially structured Excel worksheet.  This presentation will be of interest to those seeking to optimize marketing campaigns of any size while managing operational and computational complexity. Attendees will be given an electronic copy of the presentation, as well as the Excel worksheet used for optimization, and instructions on how to load the Solver Add-in.

3:00 – 3:30     Break
3:30 – 4:15    Advances in Modeling Dynamics II
Incorporating Experience Quality Data into CRM Models: Scalable Estimation of Dynamic Learning Models with Amazon Elastic Cloud Computing.

Wayne Taylor and Anand Bodapati – Southern Methodist and UCLA

In many industries, the quality of a customer’s experience may fluctuate across transaction occasions. We propose a framework and methodology to model a customer’s evolving believes about the firm’s quality and how these beliefs combine with marketing to influence purchase behavior. Our methodology combines classic econometric approaches with state-of-the-art Amazon EC2 cloud computing technology to enable estimation of a previously intractable learning model. This research allows managers to assess the marketing response of a customer with any specific experience and behavior history, which can be used to decide which customers to target for marketing.

4:15 – 5:00    Round-Table Discussions
5:00 - 6:30     Reception
6:30 - 8:00    Buffet dinner and entertainment by Brock Dotson

Friday, June 22, 2018

7:00 – 8:20    Breakfast
8:20 – 8:30    Introduction to Session – Jeff Dotson, BYU
8:30 – 10:00    New Frontiers I
The Advantages of Substituting a Rating/Ranking Task for Maximum Difference Scaling (Max-Diff) with a List Containing a Large Number of Inapplicable Items
Ben Cortese and Larry Goldberger – KS&R Inc.
While Max-Diff has proven to be a highly robust and valuable technique to determine the relative importance of a list of items, problems arise when some of the items being tested are irrelevant to some of the respondents in the population being studied. In the first study we use a split sample design which illustrates the advantages of using a rating and ranking task (positive Q-sort) as an alternative to Max-Diff in this situation.  While the Max-Diff task resulted in dropout rates of over 50% in most countries, the rating/ranking task reduced the dropout rate to under 20%. In a second study we contrast two ways of analyzing inapplicable items using Hierarchical Bayes: 1) Treating them as unimportant and thus as the lowest ranked items or 2) Excluding the items from the analysis on a case by case basis.  The results show that the two analytic techniques produce similar aggregate results but the exclusion technique is far superior in capturing individual differences.  This later finding suggests that HB may not be very robust in capturing unsmooth distributions. 

The Role of Community Feedback in Online Idea Contests
Daniel Zantedeschi – Ohio State

Online idea contests are today an increasingly popular approach for organizations to leverage the creativity of the crowd for new product development. In these contests, ideators present their ideas to an online community of potential users, who often provide feedback for idea improvement or refinement. However, it is unclear if and under what conditions this feedback helps improve idea quality. This study examines the direct and moderating effects of community feedback on idea quality in such environments. Using data from crowdsourcing contests sponsored by Zeiss and Fujitsu, our empirical analysis suggests that, perhaps counterintuitively, more feedback may have detrimental effects on idea quality. Moreover, the number of users providing feedback plays an important moderating role. Based on the empirical estimates and counterfactual exercises, we show that increases in users and learning-driven feedback affects idea quality negatively, while increases in users and task-driven information have positive effects. Implications of our findings for theory and practice of marketing research are discussed as well as general recommendations for better designing and monitoring innovation contests.

10:00 – 10:30    Break
10:30 – 11:15    New Frontiers II

Can User Generated Content Predict Restaurant Survival: Deep Learning of Yelp Photos and Reviews?
Mengxia Zhang and Lan Luo – University of Southern California

We use deep learning methods to analyze 795,175 photos and 1,015,825 reviews posted on Yelp from 2004 to 2015 on 17,796 restaurants. Tracking the survival of these restaurants during this time period, we find that both volume and valence of photos are strong predictors of restaurant survival. Nevertheless, when it comes to reviews, only valence (not volume) matters. Our research is among the earliest attempts to introduce both photo- and text- based deep learning in marketing. We are also the first to compare and contrast managerial impacts of consumer posted photos vs. reviews. To our knowledge, this is also the first large-scale empirical research on restaurant survival.

11:30    Adjournment and Lunch – Greg Allenby

Call for Presentations: Deadline is February 12, 2018

The 29th ART Forum features practitioner and academic presentations on techniques useful to marketing researchers.  It encompasses traditional topics such segmentation, targeting and positioning and embraces new frontiers of text data, field experiments and big data analytics for improving business performance.  Practitioners coming to the ART Forum will learn about new solutions and academics  will learn about new problems and practitioners will learn about new solutions. 

The ART Forum has historically been a rich venue to advance the science and practice of marketing thought.The program committee for this year's conference is Greg Allenby from Ohio State, Jeff Dotson from BYU and Elea Feit from Drexel University.  They encourage you to submit proposals for presentation for this year's conference.  In keeping with past ART Forum conferences, we continue to value solutions and source code that can be shared among participants (i.e., so sales presentations, please!).  We look forward to hear from you.

Greg Allenby, Conference Chair


Presentations may be scheduled for 15-45 minutes as the committee finds appropriate. A proposal should:

  • Demonstrate an innovative and methodologically rigorous approach;

  • Integrate a marketing problem with one or more analytic methods;

  • Clearly communicate methods and practical benefits;

  • Provide explicit and useful conclusions and recommendations for practitioners and academics.

Papers must be legitimate research offerings, and not sales presentations or pitches for proprietary methods.



Please submit your presenation/paper proposals using the following outline, describe your paper in 800 words maximum (total for items 4 – 7).

1. Title of the proposed paper.

2. Names of authors.

3. Specification of the research methods and techniques used (type of data, modeling approach, etc.).

  • What are the data sources (simulated data, choice model observations, survey, market basket, etc.)?
  • With respect to analytic methods or programs: are they (1) commercially available, (2) privately available (consultancy proprietary or custom programming), or (3) available as open/shared source?

4. Status of Research: Proposed, underway or complete (including forthcoming/recently published).

5. Relevance and Importance: Reasons why the proposed paper will be of interest and practical value to the ART Forum audience of advanced research practitioners and academics.

6. Take-Home Materials: Materials that will be provided as guides for conference attendees to implement the methods. Examples of potential take-home materials includes working papers, sample code, and/or how-to implementation guides. Will you share your code or procedure, plus example data (synthetic data is OK), such that attendees can easily recreate an analysis of the type you demonstrate? (These materials are not required at time of proposal submission, but will be required with the draft presentation.)

7. Presentation Format: (1) Research presentation; (2) Software, Code, or Method demonstration; (3) Poster.

8. Requested Amount of Time: 15 minutes (brief paper or demonstration), 30 minutes (extensive paper), 45 minutes (keynote or review paper), or NA (poster).

9. References: If available, please list (selected) earlier work by the authors to provide a context for the contribution; if applicable, please list other conferences where this paper has been presented.

10. Promotional description of your presentation (4-5 sentences) that will be used in program materials if accepted.

Please send submissions by February 12, 2018 as .DOC files with the subject line: 2018 ART Forum Submission to: Matt Weingarden, mweingarden@ama.org



The 2016 Program> is a good guide to the kinds of papers that appear at ART Forum. Additionally, here is a quick list of papers that have been honored as top papers at past events:
  • “Buy Till You Die” to Improve Operational Efficiency: Realigning the Sales Force Using the BG/NBD. Steven Lerner, Peter Fader, Bruce Hardie (2012)
  • Consumer Preference Elicitation of Complex Products using Support-Vector-Machine (SVM) Active Learning. Lan Luo, Dongling Huang (2012)
  • Developing an Advanced Preference Based Simulator – A Flexible and Scalable Approach Using Rand the Amazon Elastic Compute Cloud. Charles Carpenter, Jeffrey Dumont, Nelson Whipple (2012)
  • Models of Sequential Evaluation in Best-Worst Choice Tasks. Tatiana Dyachenko, Rebecca Walker Naylor, Greg M. Allenby (2013)
  • Monitoring Shifts in Product Feature Importance via Trends in Online Searches. Ye Hu, Rex Yuxing Du, Sina Damangir (2014)
  • Beyond Pairwise Similarity: The Category Covering Problem for the Analysis of Sorting Task Data in Marketing Research. Simon J. Blanchard, Wayne S. DeSarbo, Daniel Aloise (2015)


Already registered on site? Please log in first in the upper right corner of the site before proceeding.

Select Attendee Type:




Select Event Tutorials:

Thursday - 12:30 PM - 2:30 PM
Non-Members : $0.00
Members : $0.00
Thursday - 3:00 PM - 5:00 PM
Non-Members : $0.00
Members : $0.00
Thursday - 12:30 PM - 2:30 PM
Non-Members : $0.00
Members : $0.00
Thursday - 3:00 PM - 5:00 PM
Non-Members : $0.00
Members : $0.00
My Selected Tutorials:
Non-Member $
Member $
Base Price For The Event

Conference Location

The Blackwell Inn and Pfahl Conference Center
2110 Tuttle Park Place
Columbus, OH 43210
Telephone: +614.247.4000
Reservations: +1.866.247.4003

A simple craftsman design aesthetic, upscale accommodations, and contemporary amenities to keep you comfortable and pampered are just a few of the reasons our hotel is unparalleled in Columbus. The Blackwell Inn and Pfahl Conference Center offers 151 elegantly designed guestrooms, 20,000 square feet of sophisticated meeting and event centers, complimentary airport transportation, and a convenient location close to downtown Columbus’ arts and entertainment district. And we’re the only hotel on The Ohio State University campus, allowing easy access to sporting venues, recreational facilities, and other campus highlights..

Conference Attire
Conference attire is business casual. Meeting rooms tend to be cool so you may wish to bring a sweater/jacket.

The AMA is committed to providing equal access to our meetings for all attendees. If you are an attendee with a disability and require program accommodations, please contact the AMA Meeting Services Department, and a member of our staff will ensure that appropriate access arrangements are made. If you have specific disability related needs for your hotel sleeping room, please be sure to communicate those directly to the hotel when you make your reservation. In an effort to provide the highest quality of service to all attendees, we require that details of all access requests be communicated to our office at least 14 days in advance of the beginning of the meeting.

AMA Travel Program

American Marketing Association has partnered with Delta to provide our attendees a 2% - 10% discount for the ART Forum in Columbus, OH. The Delta Meeting Network discount is based on the booking class of your ticket.

Reserve your ticket now​ 
or call Delta Meeting Network Reservations at +1.800.328.1111.  When booking online, select Book A Trip, click on Advanced Search and enter the meeting event code: NMPS2 in the box provided on the Find Flights page.

Reservations may also be made by calling Delta Meeting reservations at +1.800.328.1111 Monday through Friday 7:00 am - 7:00 pm CDT.
*Please note that a Direct Ticketing Charge will apply for booking by phone through the reservation number above.

Tower Travel Management
AMA's travel coordinator, Tower Travel Management, is available to assist with reservations. Call +1.800.542.9700 within the U.S. and Canada. Reservation lines are open Monday through Friday 7:00 am - 7:00 pm CDT or, you may contact them via email at association@towertravel.com.

Tower Travel will proactively research airfares on ALL airline carriers to ensure that the lowest available fares are offered to all attendees. They are dedicated to providing superior customer service and hassle-free travel arrangements. Please note that fees, restrictions and cancellation penalties will apply.

​Sponsor & Exhibition Opportunities: 

To learn more about AMA Sponsorship opportunities, please contact Andrew Nelmes at anelmes@ama.org or call 312.542.9088.​

Customized partnerships are also available!

Become a Member
Access our innovative members-only resources and tools to further your marketing practice.

 Upcoming Webcasts


 Why Attend


"The ART Fourm has helped me gain proficiency in hierarchical Bayes methods, segmentation, and probability models.  I apply these methods in every project I conduct!"
Past ART Forum Attendee

"The ART Forum helped me develop from a textbook statistician to a creative market researcher."
Past ART Forum Attendee

 "The ART Forum provides a great opportunity to network with marketing research practitioners that are working on interesting problems."
Past ART Forum Attendee


 Join the AMA

​The AMA connects you to a world of resources that deliver results, and help you succeed today and into the future. Join the AM, and put the power of AMA membership to work for you.