Dissimilarity profile analysis (DPA) is a multivariate exploratory tool that extends the usual profile analysis for data matrices to dissimilarity matrices. Main feature of DPA is the decomposition of the squared Euclidean distance between pairs of dissimilarity profiles in four additive components, each of which has a proper meaning. Potentialities of DPA are illustrated with reference to a case study concerning a missing data imputation problem in the context of cardiovascular risk assessment.
Solaro, N. (2017). Dissimilarity profile analysis for assessing the quality of imputation in cardiovascular risk studies. In Cladag 2017 Book of Short Papers. Mantova (MN) : Universitas Studiorum S.r.l. Casa Editrice.
Dissimilarity profile analysis for assessing the quality of imputation in cardiovascular risk studies
Solaro, N
Primo
2017
Abstract
Dissimilarity profile analysis (DPA) is a multivariate exploratory tool that extends the usual profile analysis for data matrices to dissimilarity matrices. Main feature of DPA is the decomposition of the squared Euclidean distance between pairs of dissimilarity profiles in four additive components, each of which has a proper meaning. Potentialities of DPA are illustrated with reference to a case study concerning a missing data imputation problem in the context of cardiovascular risk assessment.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
