Academic

Missing data occur frequently in empirical studies in health and social sciences, and can compromise our ability to obtain valid inference. In missing not at random (MNAR) settings, identification is generally not possible without imposing additional assumptions. In this talk, we provide necessary and sufficient conditions for nonparametric identification of the full data distribution under MNAR with the aid of a valid instrumental variable. For inference, we focus on estimation of a population outcome mean, for which we develop a suite of semiparametric estimators that extend methods previously developed for data missing at random. For illustration, the methods are used to account for selection bias induced by HIV testing refusal in the evaluation of HIV seroprevalence in Mochudi, Botswana, using interviewer characteristics such as gender, age and years of experience as instruments.