Andreas Weigend
Stanford University
Stat 252 and MS&E 238

Data Mining and Electronic Business

Homework 4: Developing an Online Consumer Confidence Index

Since 1985, the Conference Board has produced a gauge of consumers' confidence in the US economy as well as 6 month expectations. This index is based on a mailed survey aiming for n=5000 returns with 5 questions:
  1. Business conditions
  2. Business conditions 6 month outlook
  3. Current employment conditions
  4. Employment conditions 6 month outlook
  5. Expectations of family’s income 6 month outlook
Respondents rate current and expected conditions by positive (+1), neutral (0) or negative (-1).
The response proportions to each question are seasonally adjusted. For each of the five questions (above), the POSITIVE figure is divided by the sum of the POSITIVE and NEGATIVE to yield a proportion, which we call the "RELATIVE" value. For each question, the average RELATIVE for the calendar year 1985 is then used as a benchmark to yield the INDEX value for that question. The Indexes are then averaged together as follows:
  • Consumer Confidence Index: Average of all five Indexes;
  • Present Situation Index: Average of Indexes for questions 1 and 3;
  • Expectations Index: Average of Indexes for questions 2, 4, and 5.
Developing an Online Consumer Confidence Index:
  • Students to pair into teams of between 2 to 5 students (no solos allowed)
  • Using historical search term data, find word-pairs, search term portfolios or term breadth that provide insight into consumer economic sentiment
  • Index should address these four areas:
    • Housing (example: ratio of homes for sale, homes for rent queries)
    • Retail
    • Employment
    • Travel
Project Timing:
  • Teams should submit a 1-2 page proposal of the mechanism for their index by Sunday May 5th. Proposals should indicate how the team will calculate the index, what terms, word pairs, volume v. breadth measures will be used in the index
  • Finalists will be notified by Sunday May 12th of the accepted proposals
  • Final index formula due by class 8
  • Index will be examined through end of year (Dec 2007) for insight into economic sentiment and reliability in predicting consumer driven indicators
  • The winning team will be notified first week in January 2008.

What You’re Playing For:
  • Winning team will be recognized and cited in any published research on the topic
  • Andreas Weigend and Bill Tancer will take the winning team to a dinner in SF (location TBD)
  • Winning team will split a $500 in American Express Gift Certificate