New Developments in Panel Data Estimation: Full-Factorial Panel Data Model

Panel data has been widely used in many social science studies. Pooling data across cross-sections and time-series improves quality of data analysis; however, the model is limited in its ability to actually accurately predict variables of interest due to severe practical data limitations and the ability of properly capturing varying market structures. In this article, a simple and innovative model of product share is introduced. The Full-Factorial Panel Data Model is based on the simple premises of re-conceptualization of any zero-sum group as a series of two-entity markets. This model solves the challenges associated with pooling data across disparate cross-sections and time-periods as well as the changing competitive market structure issues and therefore results in reliable variable of interest estimates.


Issue Date:
2007
Publication Type:
Conference Paper/ Presentation
PURL Identifier:
http://purl.umn.edu/9355
Total Pages:
13
Series Statement:
Selected Paper 171417




 Record created 2017-04-01, last modified 2017-08-22

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)