|
Activity Number:
|
70
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistics and Marketing
|
| Abstract - #303367 |
|
Title:
|
Implicit Regression Modeling in Supercritical Pitchfork Bifurcation Approach
|
|
Author(s):
|
Stan Lipovetsky*+
|
|
Companies:
|
GfK Custom Research North America
|
|
Address:
|
8401 Golden Valley Rd., Minneapolis, MN, 55427,
|
|
Keywords:
|
Chaos ; supercritical pitchfork bifurcation ; regression of implicit function ; marketing and advertising research
|
|
Abstract:
|
Regression modeling by messy data is considered in so called Supercritical Pitchfork Bifurcation (SPB) approach known in description of chaotic systems. The work describes a possibility of applying SPB technique to regression modeling of the implicit functions, so to construct a non-unique response dependency. Theoretical properties and practical advantages of SPB regression are discussed with an example from marketing research data on advertising in the car industry. The results show that the SPB model is very promising in modeling, analysis, interpretation, and leads to better understanding of complicated data. The suggested approach can help researchers make more adequate regression assessments for solving practical problems.
|
