A Proposed Generalized Method of Moment Estimation of the Panel Vector Autoregressive Model with Missing Data

Abstract

Estimation methods to handle missing data problems in various panel data models are rarely discussed. However in the panel vector autoregressive model, there is no estimator to handling this type of problems. The traditional treatment for cases of involving incomplete data is to use generalized method of moment estimation based only on the available data without imputation the missing data. Thus, this study introduces a generalized method of moment estimator for the panel vector autoregressive model for solving problems involving incomplete dataset. Our proposed estimator is based on mean imputation with generalized method of moment estimation to achieve more efficiency. The statistical properties of the proposed estimator are presented. Additionally, real data application is used to investigate the efficiency of the proposed estimator compared with the traditional generalized method of moment estimator. The results showed that the proposed estimator is better than the traditional generalized method of moment estimator, even if the percentage of missing data is up to 40%.