Using gret l for Principles of Econometrics, 4th Edition

Pooling Time-Series and Cross-Sectional Data

A panel of data consists of a group of cross-sectional units (people, firms, states or countries) that are observed over time. Following Hill et al. (2011) we will denote the number of cross-sectional units by N and the number of time periods we observe them as T.

In order to use the predefined procedures for estimating models using panel data in gretl you have to first make sure that your data have been properly structured in the program. The dialog boxes for assigning panel dataset structure using index variables is shown below:

Panel data (stacked time series)

716 cross-sectional units observed over 5

о iUse index variab es

To use this method, the data have to include variables that identify each individual and time period. Select the Panel option using the radio button and gretl will then be able to work behind the scenes to accurately account for the time and individual dimensions. The datasets that come with this manual have already been setup this way, but if you are using your own data you'll want to to assign the proper dataset structure to it so that all of the appropriate panel data procedures are available for use.

Gretl gives you easy access to a number of useful panel data sets via its database server.[82] These include the Penn World Table and the Barro and Lee (1996) data on international educational

attainment. These data can be installed using File>Databases>On database server from the menu bar as shown in Figure 15.1 below. From here, select a database you want. In Figure 15.2

On database server..

Figure 15.1: Accessing data from the database server via the pull-down menus

the entry for the Barro-Lee panel is highlighted. To its right, you are given information about whether that dataset is installed on your computer. Double click on barro_lee and a listing of the series in this database will appear in a new window. From that window you can search for a particular series, display observations, graph a series, or import it. This is a VERY useful utility, both for teaching and research and I encourage you to explore what is available on the gretl server. You will notice the 4 icons at the top of the window

The first icon from the left is the list series icon. Clicking it will bring up the list of series in a particular database as shown below in Figure 15.3. The icon that looks like a floppy disk (remember those?) will install the database. The clicking the next icon will show which databases are installed on your computer, and the ‘X’ closes the window.

Figure 15 shows a portion of the series list window for the Barro and Lee data from the database server. From here you can display the values contained in a series, plot the series, or add a series to your dataset. Highlight the particular series you want and click on the appropriate icon at the top.