Panel data refers to multi-dimensional data frequently involving measurements over time in econometrics. It is also called longitudinal data in biostatistics.
Panel data (also called longitudinal data) consist of data that are collected repeatedly on the same study units (e.g., firms or subjects). This type of data allows one to exploit both cross-sectional and time series information on the sampled subjects. This makes it possible to eliminate endogeneity problems due to unobserved factors which are invariant over time. Such fixed effects can be absorbed or differenced out (see fixed effects estimation). If such effects are of no concern, it is possible to improve on OLS in terms of efficiency by using the random effects estimator which utilizes the between and within information in the data more effectively.
Many estimation techniques rely on so-called "small T large N" asymptotic, i.e. many subjects or series that are observed for a relatively short time period. As the time dimension increases, the data becomes more dynamic, leading to inconsistencies in the standard panel estimators. Methods for dealing with dynamic panel data have been developed by Anderson and Hsiao, and Arellano and Bond, among others.
Examples of longitudinal data sets include the Panel Study of Income Dynamics (PSID), the British Household Panel Survey (BHPS) or the National Longitudinal Survey (NLS).
For an extensive overview of panel econometric and statistical techniques see for instance:
Wooldridge, J. (2010). Econometric Analysis of Cross Section and Panel Data (2nd ed.). Cambridge, MA: MIT Press.