Incomplete Data Analysis for Economic Statistics

Abstract

Incomplete data are ubiquitous in social sciences; as a consequence, available data are inefficient and often biased. This dissertation deals with the problem of missing data in official economic statistics. Building on the practices of the United Nations Economic Commission for Europe (UNECE), the first half of the dissertation focuses on single imputation methods. After revealing that single ratio imputation is often used for economic data in the current practices of official statistics, this study unifies the three ratio imputation models under the framework of weighted least squares and proposes a novel estimation strategy for selecting a ratio imputation model based on the magnitude of heteroskedasticity. After showing that multiple imputation is suited for public-use microdata, the latter half of the dissertation focuses on multiple imputation methods. From a new perspective, this dissertation compares the three computational algorithms for multiple imputation: Data Augmentation (DA), Fully Conditional Specification (FCS), and Expectation-Maximization with Bootstrapping (EMB). It is found that EMB is a confidence proper multiple imputation algorithm without between-imputation iterations, meaning that EMB is more user-friendly than DA and FCS. Based on these findings, the current study proposes a novel application of the EMB algorithm to ratio imputation in order to create multiple ratio imputation, the new multiple imputation version of ratio imputation, providing brand-new software MrImputation implemented in R. Combining all of these findings, this dissertation will be an important addition to the literature of missing data analysis and official economic statistics.