Many relevant multidimensional phenomena are defined by nested latent concepts, which can be represented by a tree-structure supposing a hierarchical relationship among manifest variables. The root of the tree is a general concept which includes more specific ones. The aim of the paper is to reconstruct an observed data correlation matrix of manifest variables through an ultrametric correlation matrix which is able to pinpoint the hierarchical nature of the phenomenon under study. With this scope, we introduce a novel model which detects consistent latent concepts and their relationships starting from the observed correlation matrix.

Additional Metadata
Keywords Ultrametric correlation matrix · Hierarchical latent concepts · Partition of, variables · Hierarchical factor models · Higher-order models
Persistent URL dx.doi.org/10.1007/s11634-020-00400-z, hdl.handle.net/1765/127248
Journal Advances in Data Analysis and Classification
Note Mathematics Subject Classification: 62H25
Citation
Cavicchia, C, Vichi, M, & Zaccaria, G. (2020). The ultrametric correlation matrix for modelling hierarchical latent concepts. Advances in Data Analysis and Classification. doi:10.1007/s11634-020-00400-z