Messy Data in Heteroscedastic Models Case study

Missing data are popular in real experiments. Messy data may give rise to heterogeneity of variance, where missing of data are dramatically different than any analyze of these data is based. Missing data may all cell is empty or loss some observation in the cell which mean unbalanced data. In this book attempts to investigate the theoretical side of nested design analysis in the case of unbalanced model in the situation where approximate methods such as unweighted means are inappropriate. Instead of estimate the missing observations; In this book author adjusted the estimating methods to deal with messy data problem.