ART2010 Tutorials

Lynn Brown
Current average rating    
 
Key Takeaways
Keywords
Sunday, June 6th, 2010
8:00am – 12:00noon                     Concurrent Tutorials A, B, C, D
 
 A.  An Introducation to R-Code for Marketing Research
Thomas Lumley, University of Washington
 
This tutorial will provide an introduction to the open-source R statistical programming environment, covering data handling, the package system, graphics, basic statistical modelling, and simple programming. Knowledge of R is useful in marketing research not only as a data analysis system, but because new statistical methods are increasingly made available as R packages. The tutorial will include practical exercises; attendees should bring a laptop and should download and install R from http://www.r-project.org/ before the session.
 
B.  Introduction to Discrete Choice Modeling
Brian Orme, Sawtooth Software and Jon Pinnell, MarketVision Research
 
This tutorial is designed for people with only limited background or experience with discrete choice modeling.  We’ll provide background for understanding from where choice models have evolved and why they have gained such presence in commercial marketing research over the past decade.  In addition, we’ll:
· Review the terms commonly encountered when designing, executing, analyzing and reporting a discrete choice study.
· Discuss the inputs to and outputs from a discrete choice study.
· Describe the challenges of modeling choice data, including IIA, capturing heterogeneity, and dealing with simple vs. more complex model specifications.
· Provide a list of ‘pitfalls to avoid’ and ‘best practices’ for executing discrete choice studies.
· Illustrate uses and applications of discrete choice modeling.
 
While some math is unavoidable, the focus is on practical issues, solutions, and theory.  We will provide attendees a framework by which they can evaluate whether discrete choice is an appropriate approach and how best to work with internal specialists or vendors to deliver the most useful choice research.
  
C.  Applied Probability Models in Marketing Research: Introduction.
Peter Fader, University of Pennsylvania and Bruce Hardie, London Business School
 
Central to a complete understanding of today’s “leading-edge” market research techniques is a sound intuitive appreciation of the basic foundations upon which these sophisticated tools are built. For example, both Hierarchical Bayes models and latent class models build on simple probability modeling concepts (e.g., zero-order choice process, Poisson counts, conditional expectations, and exponential interpurchase times)—yet how many researchers are comfortable at precisely defining these concepts or explaining the motivation for using them?
 
This tutorial aims to fill in these gaps by bringing practitioners fully up to speed on the basic methods that may underlie many of their current or future research activities. Our two broad objectives are (1) to review the basic terminology and logic associated with the area of probability models as applied to marketing research problems, and (2) to develop participants’ skills through a set of case studies that demonstrate the model building process in detail. We will illustrate all of the steps required to develop a probability model, estimate its parameters, and interpret the results. Careful and extensive use is made of the Solver tool in Microsoft Excel, which makes it possible to construct all of these models within a familiar spreadsheet environment. By the end of the tutorial, participants should be quite comfortable with all of the aforementioned principles and models and the managerial issues that surround them
 
D.  Market Segmentation: Conceptual and Methodological Foundations
Wagner Kamakura, Duke University
 
Market segmentation is an essential component of any marketing strategy, and is a required consideration in most marketing-related decisions.  As many organizations become more customer-focused, they also use segmentation as an important basis for developing their customer relationship-management strategy.
 
This tutorial will start with the conceptual foundations of Market Segmentation, reviewing the requirements for effective segmentation, and the different forms of market segmentation. However, the major emphasis of this tutorial will be in providing participants with a clear intuition about how the basic methods for market/customer segmentation work, what are their advantages and limitations.  This will be done through the discussion of illustrative examples and real applications of the methodology for market and customer segmentation based on life-style, life-cycle, choice-based conjoint, customer behavior and share-of-wallet.  Among the methods and models to be discussed are: K-means Clustering, Latent-class Analysis, Regression Mixtures and Multinomial-Logit Mixtures.
 
This tutorial will draw on the material from the book of the same title by Wedel and Kamakura, but with less emphasis on the technical details, and with new methods and applications.
 
 
1:00pm – 5:00pm                       Concurrent Tutorials E, F, G, H
 
E.  Making the Most of the ART Forum: An Exploration of Topics and Methods
Jeff Brazell and Lauren Kippen, The Modellers, Inc.
 
Get ahead of everyone else! If you would like some extra background on and discussion of the important advancements that will be presented at this year’s ART Forum, then this tutorial is for you. We will outline and discuss the major analytical themes which will be covered in this year’s Forum with the aim of helping you get the most out of the ART.  We’ll include some history to help put methodologies into perspective and incorporate case study examples where possible.  Topics will follow the Forum agenda and include customer relationship modeling, choice modeling, dynamic models, as well as data collection challenges and implications
  
F. Probability Models for Customer Base Analysis
Peter Fader, University of Pennsylvania
Bruce Hardie, London Business School
 
Customer-base analysis seeks to use information on the history of customer purchase patterns to identify which individuals are most likely to be active (or inactive) customers and to predict future purchasing patterns by those customers listed in the firm’s transaction database. Any researcher hoping to make statements about “customer lifetime value” must deal with these issues, but unfortunately the set of commonly available tools is not well-suited for the task.
 
This tutorial builds upon the basic “platform” provided in our introductory seminar to provide a set of techniques and models tailored to address these situations properly. As before, we will focus on developing the models entirely in Excel and provide attendees with the relevant spreadsheets and notes on how to implement the models “from scratch”. Our goal is to provide the attendee with tools that can be applied immediately (maybe with some slight modifications) at his/her place of work. The structure of the tutorial is as follows:
· Introduction to the idea of customer-base analysis
· Overview of the concept of Customer Lifetime Value (CLV) and the presentation of a general framework for its calculation
· Brief review of the probability modeling basics required for model building (e.g., review of binomial, geometric, Poisson, exponential, gamma, and beta distributions; discussion of common mixtures such as the NBD, beta-geometric, and beta-binomial)
· Presentation of probability models that can be used to answer various managerial questions including the calculation of CLV
· Generalizations of the specific models presented in this tutorial making links to the broader modeling literature
 
G.  An Introduction to Bayesian Statistics and Marketing
Greg Allenby, The Ohio State University
 
The tutorial will provide an introduction to modern Bayesian statistics in Marketing.  The tutorial will draw from material in the book Bayesian Statistics and Marketing by Rossi, Allenby and McCulloch using the R statistical software available at http://www.r-project.org/.  Many sessions at the conference this year will report results based on (hierarchical) Bayesian models using Markov chain Monte Carlo (MCMC) methods of estimation.  The goal of this tutorial is to lay the groundwork for understanding the basics behind these sessions, and to set the stage for the advanced tutorial offered on Wednesday.  Software algorithms from the R contributed package "bayesm" will be used to illustrate the methods (also available from the same website).  Attendees are encouraged to download R and the bayesm package onto their laptops before the session, and to bring their laptops with them to the tutorial.  Attendees will leave with a better understanding of Bayesian models and some of the tools needed to estimate their own models of marketing data.
   
H.  Advanced Market and Customer Segmentation
Wagner Kamakura, Duke University
 
This second tutorial will focus on extensions of the basic classification and data-reduction tools discussed in the first tutorial, which allow for more advanced forms of market and customer segmentation such as: segmentation with concomitant variables, joint segmentation and dynamic segmentation.
 
To better understand these model extensions, we will first review some of the methodological details of the basic models.  Among the extensions to be covered in the advanced tutorial are:
· Latent class analysis with concomitant variables – useful for profiling segments and for joint segmentation
· Hidden Markov models - applied to dynamic segmentation
· Latent variable models for Generalized Factor Analysis – useful for data reduction and for individual-level measurement
         
Implementation issues will also be discussed in this second tutorial, including the availability of commercial software for latent-class and latent-variable modeling
 
Wednesday, June 9th, 2010
 
1:00pm – 5:00pm                       Concurrent Tutorials J, K, L
 
J.  Introduction to Dynamic Marketing Models
Prasad Naik, University of California, Davis
 
Dynamics emerge in many ways: managers’ current decisions (e.g., advertising spending) affect not only current outcomes (e.g., sales, awareness), but also future outcomes (e.g., brand equity); or managers anticipate future outcomes (e.g., likelihood of product harm crisis) and incorporate potential consequences in making current decisions; even the mere passage of time changes a firm’s external environment (e.g., onset of recessions), affecting consumers and managers and competitors and markets. Although such different phenomena lead to different formulations of dynamic models, their estimation and inference is, nonetheless, unified in the framework of state-space models.
This tutorial will introduce several examples of commonly-used dynamic models in marketing, illustrate the common theme that unifies them within the state-space framework, explain how to estimate them using a common algorithm via the Kalman filter, and provide an intuitive understanding of how and why the Kalman filter works and its role in classical and Bayesian statistics. This tutorial contains technical notations for clarity and precision (viewer discretion is advised!), but our focus will be on learning the use and usefulness of dynamic marketing models in practice.
 
K.        Advanced Topics in Discrete Choice Modeling
Jon Pinnell, MarketVision Research and Brian Orme, Sawtooth Software
 
In this tutorial, we’ll discuss advanced topics the experienced choice researcher is likely to encounter in practice.  This session will discuss issues surrounding design, parameter estimation, and simulation.  The session will begin with very brief review of background and definitions.  The session will continue with:
· Aanced design topics, such as evaluating design efficiency, alternative-specific designs, interactions, conditional effects, and methods to deal with many attributes;
· Issues in parameter estimation, such as different approaches to disaggregation, HB pitfalls, assessing respondent reliability; the scale parameter, methods to guide market segmentation; and
· Simulation topics, including: the question if utilities are linearly additive, goal seeking/optimization, IIA issues and interpretation of the default alternative. 
 
Additional topics will be included such as maximum difference models and data fusion.  The session will include a compendium of ‘best practices’ based on empirical findings.  We assume that those attending this tutorial are familiar with discrete choice modeling.
 
L.  R-Code for Marketing Research (Repeat)
Thomas Lumley, University of Washington

 


Author Bio:

 
Lynn Brown
Add A Comment :