New methods for generating significance levels from multiply-imputed data

Více o knize

The dissertation addresses the pervasive issue of missing data in statistical analyses, highlighting the significance of multiple imputation as a solution. Initially proposed by Rubin, this technique has evolved and gained prominence due to advancements in computing power over the last decade. The work emphasizes the importance of multiple imputation in applied statistics, showcasing its effectiveness in managing missing values across various settings. The research is recognized for its high academic quality, achieving a summa cum laude distinction.