Zero Replacement in Compositional Data Sets

摘要: The sample space of compositional data is the open simplex. Therefore, zeros in a set are identified either with below detection limit values, or lead to division into different subpopulations corresponding lower dimensional space. Most multivariate analysis techniques require complete matrices, thus calling for strategy imputation first case. Existing replacement methods rounded reviewed, and new method proposed, who’s properties analyzed illustrated. applied hierarchical cluster data.