Panel estimation of state dependent adjustment when the target is unobserved

Understanding adjustment processes has become central in economics. Empirical analysis is fraught with the problem that the target is usually unobserved. This paper develops, simulates and applies GMM methods for estimating dynamic adjustment models in a panel data context with partially unobserved targets and endogenous, time-varying persistence. In this setup, the standard first difference GMM procedure fails. I propose three estimation strategies. One is based on quasi-differencing, and it leads to two different, but related sets of moment conditions. The second is characterised by a statedependent filter, while the third is an adaptation of the GMM level estimator.