Roderick J.A. Little and Donald B. Rubin (2002) Statistical Analysis with Missing Data, 2nd ed. John Wiley & Sons

Abstract

Almost two decades since its release, Little and Rubin’s Statistical Analysis with Missing Data,
remains a timeless resource for anyone analysing or wanting to do research for incomplete data. In
particular, its emphasis on underlying patterns and mechanisms, and simultaneous presentation of
various inferential tools, facilitate choosing appropriate methods for analysis, and enable accurate
comparison of approaches. Its reference list provides the curious reader a great resource on the
classical literature and allows him/her to navigate through the recent research. In the era of big data,
more and messier data will increasingly become available. Such increased volume and variety of
missing data necessitates analysts to understand and correctly choose appropriate methods, and
for researchers to develop new methods when existing methods fail.