Browse/search for people

Publication - Dr Chris Muris

    Efficient GMM estimation with incomplete data

    Citation

    Muris, C, 2019, ‘Efficient GMM estimation with incomplete data’. Review of Economics and Statistics.

    Abstract

    The standard missing data model classifies data in terms of “binary missingness”, that is, as either complete or completely missing. Thus, the model deals with two strata of missingness. However, applied researchers face situations with an arbitrary number of strata of incompleteness. Examples include unbalanced panels, and instrumental variables settings where some observations are missing some instruments, and other observations are missing different instruments. In this paper, I propose a model for settings where observations may be incomplete, with an arbitrary number of strata of incompleteness. I derive a set of moment conditions that generalizes those in Graham (2011) for the standard missing data setup with two strata. I derive the associated efficiency bound and propose estimators that attain it. Incompleteness is qualitatively different from binary missingness. In particular, I show that identification can be achieved even if it fails in each stratum of incompleteness.

    Full details in the University publications repository