User’s Guide : Panel and Pooled Data : Panel Estimation
Panel Estimation
EViews allows you to estimate panel equations using linear or nonlinear squares or instrumental variables (two-stage least squares), with correction for fixed or random effects in both the cross-section and period dimensions, AR errors, GLS weighting, and robust standard errors. In addition, GMM tools may be used to estimate most of the these specifications with various system-weighting matrices. Specialized forms of GMM also allow you to estimate dynamic panel data specifications. Note that all of the estimators described in this chapter require a panel structured workfile ( “Structuring a Panel Workfile”).
We begin our discussion by briefly outlining the dialog settings associated with common panel equation specifications. While the wide range of models that EViews supports means that we cannot exhaustively describe all of the settings and specifications, we hope to provide you a roadmap of the steps you must take to estimate your panel equation.
More useful, perhaps, is the discussion that follows, which follows the estimation of some simple panel examples, and describes the use of the wizard for specifying dynamic panel data models.
A background discussion of the supported techniques is provided in “Estimation Background” in ”Pooled Estimation”, and in “Estimation Background”.