Missing data: the impact on multivariate goodness-of-fit tests

Abstract

Missing data is one of the most commonly met problems in data analysis . Therefore it has attracted the attention of researchers from different scientific fields. The main focus of the research has been on developing adequate imputation procedures whose quality has been accessed in the context of predictive properties of different models of interest. However, the impact of missingness on the data distribution has not been studied thoroughly in the context of testing goodness-of-fit with multivariate distributions. In this talk, we present some of the recent developments in this direction and some new open questions regarding the topic.