Massive Choice Data

1
Massive Choice Data

• 7th Triennial Choice Symposium
• Wharton Business School
• June 13 -17, 2007

2
Impetus for Massive Data

• Technological advances (Internet, RFID)
• Computing advances
• Methodological advances
• Detailed data
• Large sample, N
• Many variables, p
• Long time-series, T
• Several products and SKUs, K

3
Goals

• Understand current state of play
• Identify issues of interest
• Review advances in models, methods, computation,
  ideas
• Discuss prospects for further research
• Any other goals that we as a group deem
  relevant

4
Outcome

• Synthesis of our deliberations to be published as
  a review paper in the Marketing Letters

5
People

• Lynd Bacon

• President, LBA Associates
• www.lba.com
• lbacon_at_lba.com

6

• Anand Bodapati

• UCLA
• anand.bodapati_at_anderson.ucla.edu

7

• Wagner Kamakura

• Duke University
• kamakura_at_duke.edu

8

• Jeffrey Kreulen

• IBM Research
• kreulen_at_almaden.ibm.com

9

• Peter Lenk

• University of Michigan
• plenk_at_umich.edu

10

• David Madigan

• Rutgers University
• dmadigan_at_rutgers.edu

11

• Carnegie Mellon University
• alm3_at_andrew.cmu.edu

• Alan Montgomery

12

• Prasad Naik

• University of California Davis
• panaik_at_ucdavis.edu

13

• Michel Wedel

• University of Maryland
• mwedel_at_rhsmith.umd.edu

14
Issues Day 1

• Session 1 (Alan)
• Computational Challenges for Real-Time Marketing
  with Large Datasets
• Session 2 (Lynd)
• Understanding Choices and Preferences with
  Massive Complex Online Data
• Session 3 (Wagner)
• Some rambling comments on High-Dimensional Data
  Analysis

15
Issues Day 2

• Session 4 (Jeffrey)
• Leveraging Structured and Unstructured
  Information Analytics to Create Business
• Session 5 (David)
• Statistical Modeling Bigger and Bigger

16
Issues Day 3

• Session 6 (Anand)
• Issues in the Modeling of Behavior in Online
  Social Networks
• Session 7 (Michel)
• State of the Art in Recommendation Systems
• Session 8 (Peter)
• Approximate Bayes Methods for Massive Data in
  Conditionally Conjugate Hierarchical Bayes Models
• Session 9 (Prasad)
• Review of Inverse Regression Methods for
  Dimension Reduction

17
Issues Day 4 (Sunday)

• Plenary Session 1
• Plenary Session 2
• Noon Adjourn