Panel Unit Root Test

Panel data with long time period have been used predominately in applied macroeconomic research like purchasing power parity, growth convergence, business cycle hypotheses, and so on. For a long panel or macro panel, time series properties will be dominating and we need to examine the stochastic behaviour of the variables. This chapter provides some theoretical issues and their application in testing for unit roots in panel data where the time dimension (T) and the cross section dimension (N) are relatively large. If N is large and T is small, the analysis can proceed only under restrictive assumptions. In cases where N is small and T is relatively large standard time series techniques are applied to systems of equations.