Abstract Details
Activity Number:
|
81
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
|
Sponsor:
|
Section on Statistics in Marketing
|
Abstract #313281
|
View Presentation
|
Title:
|
Optimal Internet Media Selection Using General Loss Functions
|
Author(s):
|
Courtney Paulson*+ and Gareth James and Lan Luo
|
Companies:
|
University of Southern California and University of Southern California and University of Southern California
|
Keywords:
|
Internet marketing ;
budget allocation ;
penalized regression ;
demand function ;
generalized lasso ;
linear constraints
|
Abstract:
|
Current methods for optimizing advertising budgets rely heavily on identifying a particular subset of advertising opportunities. However, in the case of Internet advertising, this is infeasible; the set of opportunities is limited only by the sheer number of websites. Even further, these websites often vary significantly by site traffic, advertising costs, and correlations in site visits. To address such challenges, we formulate a procedure for automatic subset and budget optimization over a very large set of Internet websites. Due to the unique nature of this problem, we optimize over a very versatile general loss function and develop an efficient algorithm for computing our optimization over a grid of tuning parameters. Furthermore, while existing methods can only handle optimal Internet selection problems on the order of 10 websites, we propose a new approach that works well for high-dimensional problems. We also demonstrate this increased dimensionality does not diminish the algorithm's efficiency. While our method performs similarly to existing methods for small numbers of websites, the proposed method can further handle budget allocation across very large numbers (e.g. 500).
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.