|
Activity Number:
|
519
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #301673 |
|
Title:
|
Functional Data Analysis for Bid Arrival Times in eBay Auctions
|
|
Author(s):
|
Shuang Wu*+ and Hans G. Müller
|
|
Companies:
|
University of California, Davis and University of California, Davis
|
|
Address:
|
5000 Orchard Park Cir 5725, DAVIS, CA, 95616,
|
|
Keywords:
|
Density function ; eBay auction ; Functional data analysis ; Intensity function
|
|
Abstract:
|
Consider the bid arrival times for online eBay auctions where one observes a relatively small number of irregularly spaced bids for each auction. We propose to model bid arrivals by postulating a random rate function as realization of an underlying stochastic process for each auction. Focusing on the random density function that reflects the shape of the rate function, a functional data analysis approach is developed by estimating mean and covariance structure of these random density functions, followed by the estimation of eigenfunctions and functional principal component scores to obtain fitted density and also rate functions for individual auctions. Theoretical results include asymptotic consistency for the components of the model. The method is illustrated with a study of bid arrival times for 156 Palm M515 PDA auctions from eBay.
|
