JSM 2011 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Abstract Details

Activity Number: 586
Type: Contributed
Date/Time: Wednesday, August 3, 2011 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics and Marketing
Abstract - #301547
Title: Tree-Based Time Varying-Coefficient Regression For Product Demand Prediction
Author(s): Jay Wang*+ and Kay-Yut Chen and Guillermo Gallego
Companies: Hewlett Packard Laboratories and Hewlett Packard Laboratories and Columbia University
Address: 1501 Page Mill Rd, Palo Alto, CA, 94304,
Keywords: elasticity ; splines ; tree-based methods ; varying-coefficient regression
Abstract:

Estimating the aggregated market demand for a product in a dynamic market is intrinsically important to manufacturers and retailers. Motivated by the need for a statistical demand prediction model to facilitate laptop pricing at Hewlett-Packard, we have developed a novel "partitioned regression" model that is easily interpretable and have good prediction power. The proposed method is generally applicable to situations where predictions are made based on a large number of categorical variables and a much smaller number of continuous variables. Similar to regression trees, the partitioned regression uses a "greedy algorithm" to find optimal partitions of the products, where "optimality" refers to the maximal reduction of $-2\log$ likelihood. Within each partition (or "market segment"), a time varying-coefficient regression model is built to incorporate the temporal dynamics of the demand-price relationship, to reflect the change of market volume and customer preference at different phases of product life cycle. The proposed methodology is applied to a real dataset containing laptop sales of all brands covering September 2009 to June 2010 in all five states of Australia.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2011 program




2011 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.