Estimating models for panel survey data under complex sampling

Complex designs are often used to select the sample which is followed over time in a panel survey. We consider some parametric models for panel data and discuss methods of estimating the model parameters which allow for complex schemes. We incorporate survey weights into alternative point estimation procedures. We also consider variance estimation using linearization methods to allow for complex sampling, and indicate connections with established asymptotically distribution free (ADF) methods. The behaviour of the proposed inference procedures are assessed in a simulation study, based upon data from the British Household Panel Survey. There appear to be some advantages of using the weighted maximum likelihood (ML) point estimation method compared to the weighted ADF method. Variance estimation methods that allow for clustering tend to lead to improvements in terms of bias. However, the variance estimator for the weighted ML estimator performs better than the ADF variance estimators.