Panel and Pooled Data

Panel and pool data involve observations that possess both cross-section, and within-cross-section identifiers.

Generally speaking, we distinguish between the two by noting that pooled time-series, cross-section data refer to data with relatively few cross-sections, where variables are held in cross-section specific individual series, while panel data correspond to data with large numbers of cross-sections, with variables held in single series in stacked form.

The discussion of these data is divided into parts. Pooled data structures are discussed first:

•
“Pooled Time Series, Cross-Section Data” outlines tools for working with pooled time series, cross-section data, and estimating standard equation specifications which account for the pooled structure of the data.

Data where the panel cross-sections are stacked are described separately:

• In
“Advanced Workfiles” , we describe the basics of structuring a workfile for use with panel data. Once a workfile is structured as a panel workfile, EViews provides you with different tools for working with data in the workfile, and for estimating equation specifications using both the data and the panel structure.

•
“Working with Panel Data”, outlines the basics of working with panel workfiles.

•
“Panel Estimation” describes least squares, instrumental variables, and GMM estimation in panel structured workfiles.

•
“Difference-in-Difference Estimation” documents describes EViews tools for the estimating panel difference-in-difference models.

•
“Panel Cointegration Estimation” documents tools for panel cointegrating regression estimation in EViews.

•
“Panel Statistics” outlines the computation of other “panel workfile aware” statistics and tests.