# Panel data

(Redirected from Longitudinal data)

In statistics and econometrics, panel data or longitudinal data[1][2] are multi-dimensional data involving measurements over time. Panel data contain observations of multiple phenomena obtained over multiple time periods for the same firms or individuals.

Time series and cross-sectional data can be thought of as special cases of panel data that are in one dimension only (one panel member or individual for the former, one time point for the latter).

A study that uses panel data is called a longitudinal study or panel study.

## Example

balanced panel: unbalanced panel:
${\displaystyle {\begin{matrix}\mathrm {person} &\mathrm {year} &\mathrm {income} &\mathrm {age} &\mathrm {sex} \\1&2001&1300&27&1\\1&2002&1600&28&1\\1&2003&2000&29&1\\2&2001&2000&38&2\\2&2002&2300&39&2\\2&2003&2400&40&2\end{matrix}}}$ ${\displaystyle {\begin{matrix}\mathrm {person} &\mathrm {year} &\mathrm {income} &\mathrm {age} &\mathrm {sex} \\1&2001&1600&23&1\\1&2002&1500&24&1\\2&2001&1900&41&2\\2&2002&2000&42&2\\2&2003&2100&43&2\\3&2002&3300&34&1\end{matrix}}}$

In the example above, two data sets with a panel structure are shown. Individual characteristics (income, age, sex) are collected for different persons and different years. In the left data set two persons (1, 2) are observed over three years (2001, 2002, 2003). Because each person is observed every year, the left-hand data set is called a balanced panel, whereas the data set on the right hand is called an unbalanced panel, since person 1 is not observed in year 2003 and person 3 is not observed in 2003 or 2001. This specific structure these data sets are in is called long format where one row holds one observation per time. Another way to structure panel data would be the wide format where one row represents one observational unit for all points in time (for the example, the wide format would have only two (left example) or three (right example) rows of data with additional columns for each time-varying variable (income, age).

## Analysis

A panel has the form

${\displaystyle X_{it},\quad i=1,\dots ,N,\quad t=1,\dots ,T,}$

where ${\displaystyle i}$ is the individual dimension and ${\displaystyle t}$ is the time dimension. A general panel data regression model is written as ${\displaystyle y_{it}=\alpha +\beta 'X_{it}+u_{it}.}$ Different assumptions can be made on the precise structure of this general model. Two important models are the fixed effects model and the random effects model.

Consider a generic panel data model:

${\displaystyle y_{it}=\alpha +\beta 'X_{it}+u_{it},}$
${\displaystyle u_{it}=\mu _{i}+v_{it}.}$

${\displaystyle \mu _{i}}$ are individual-specific, time-invariant effects (for example in a panel of countries this could include geography, climate etc.) which are fixed over time., whereas ${\displaystyle v_{it}}$ is a time-varying random component.

If ${\displaystyle \mu _{i}}$ is unobserved, and correlated with at least one of the independent variables, then it will cause omitted variable bias in a standard OLS regression. However, panel data methods, such as the fixed effects estimator or alternatively, the First-difference estimator can be used to control for it.

If ${\displaystyle \mu _{i}}$ is not correlated with any of the independent variables, ordinary least squares linear regression methods can be used to yield unbiased and consistent estimates of the regression parameters. However, because ${\displaystyle \mu _{i}}$ is fixed over time, it will induce serial correlation in the error term of the regression. This means that more efficient estimation techniques are available. Random effects is one such method: it is a special case of feasible generalized least squares which controls for the structure of the serial correlation induced by ${\displaystyle \mu _{i}}$.

### Dynamic panel data

Dynamic panel data describes the case where a lag of the dependent variable is used as regressor:

${\displaystyle y_{it}=\alpha +\beta 'X_{it}+\gamma y_{it-1}+u_{it},}$

The presence of the lagged dependent variable violates strict exogeneity, that is, endogeneity may occur. The fixed effect estimator and the first differences estimator both rely on the assumption of strict exogeneity. Hence, if ${\displaystyle \mu _{i}}$ is believed to be correlated with one of the independent variables, an alternative estimation technique must be used. Instrumental variables or GMM techniques are commonly used in this situation, such as the Arellano–Bond estimator.