Likelihood inference and the role of initial conditions for the dynamic panel data model

Abstract

Lancaster (2002) proposes an estimator for the dynamic panel data model with homoskedastic errors and zero initial conditions. In this paper, we show this estimator is invariant to orthogonal transformations, but is ine cient because it ignores additional information available in the data. The zero initial condition is trivially satis ed by subtracting initial observations from the data. We show that di erencing out the data further erodes e ciency compared to drawing inference conditional on the rst observations. Finally, we compare the conditional method with standard random e ects approaches for unobserved data. Standard approaches implicitly rely on normal approximations, which may not be reliable when unobserved data is very skewed with some mass at zero values. For example, panel data on rms naturally depend on the rst period in which the rm enters on a new state. It seems unreasonable then to assume that the process determining unobserved data is known or stationary. We can instead make inference on structural parameters by conditioning on the initial observations.