Panel-data (or longitudinal-data) methods have been central to empirical research in economics, as well as other fields in the social sciences and natural sciences. Utilizing a correlated-random effects framework for modeling panel data, the proposed research will provide several new approaches for the analysis of such data. This proposal considers new methods for estimation and inference in panel-data models, both linear and non-linear, with slope heterogeneity (in addition to the intercept heterogeneity allowed by the classical panel-data model used in economics). While fixed-effects approaches to the linear model with slope heterogeneity do exist, this research will extend the correlated random effects framework to this model, examine the equivalence of the two approaches, and discuss the usefulness of the new framework for specification testing. Far less attention has been paid to non-linear models with slope heterogeneity, despite the large literature dealing with intercept heterogeneity for these models. The proposed research will provide several new estimators for both fixed-effects and correlated-random-effects for such non-linear models. In addition to consideration of slope-heterogeneity models, the proposed research will also provide empirical approaches for two data limitations (dependent-variable misclassification and dependent-variable absence) that arise in panel-data models with intercept heterogeneity.

Broader impacts resulting from the proposed activity: The proposed research should have both a theoretical and practical impact upon economics and other fields that utilize panel-data methods. During the course of the proposed research, the PI develops computer routines for use in statistical software. Many of the proposed empirical approaches are implemented in Stata, the most widely used statistical software by applied researchers in economics. All relevant computer routines will be made publicly available in a timely manner so that empirical researchers can utilize these approaches for their work. The proposed research is also disseminated via seminar/conference presentations and doctoral student instruction.

Agency
National Science Foundation (NSF)
Institute
Division of Social and Economic Sciences (SES)
Type
Standard Grant (Standard)
Application #
0921208
Program Officer
Nancy A. Lutz
Project Start
Project End
Budget Start
2009-09-01
Budget End
2012-08-31
Support Year
Fiscal Year
2009
Total Cost
$157,320
Indirect Cost
Name
University of Texas Austin
Department
Type
DUNS #
City
Austin
State
TX
Country
United States
Zip Code
78712