Identifiability of Hierarchical Latent Attribute Models
From MaRDI portal
Abstract: Hierarchical Latent Attribute Models (HLAMs) are a family of discrete latent variable models that are attracting increasing attention in educational, psychological, and behavioral sciences. The key ingredients of an HLAM include a binary structural matrix and a directed acyclic graph specifying hierarchical constraints on the configurations of latent attributes. These components encode practitioners' design information and carry important scientific meanings. Despite the popularity of HLAMs, the fundamental identifiability issue remains unaddressed. The existence of the attribute hierarchy graph leads to degenerate parameter space, and the potentially unknown structural matrix further complicates the identifiability problem. This paper addresses this issue of identifying the latent structure and model parameters underlying an HLAM. We develop sufficient and necessary identifiability conditions. These results directly and sharply characterize the different impacts on identifiability cast by different attribute types in the graph. The proposed conditions not only provide insights into diagnostic test designs under the attribute hierarchy, but also serve as tools to assess the validity of an estimated HLAM.
Cites work
- A sparse latent class model for cognitive diagnosis
- Attribute hierarchy models in cognitive diagnosis: identifiability of the latent attribute space and conditions for completeness of the Q-matrix
- Cluster analysis for cognitive diagnosis: theory and applications
- Defining a family of cognitive diagnosis models using log-linear models with latent variables
- Estimating multiple classification latent class models
- Finite mixture models
- Fusion, propagation, and structuring in belief networks
- Hierarchical diagnostic classification models: a family of models for estimating and testing attribute hierarchies
- Higher-order latent trait models for cognitive diagnosis
- Identifiability of diagnostic classification models
- Identifiability of parameters in latent structure models with many observed variables
- Identifiability of restricted latent class models with binary responses
- Identifying Latent Structures in Restricted Latent Class Models
- Learning attribute patterns in high-dimensional structured latent attribute models
- Learning latent tree graphical models
- Model-Based Clustering, Discriminant Analysis, and Density Estimation
- On the identifiability of diagnostic classification models
- Partial identifiability of restricted latent class models
- Statistical analysis of factor models of high dimension
- Structured variable selection and estimation
- Sufficient and necessary conditions for the identifiability of the Q-matrix
- Tensor decompositions for learning latent variable models
- The DINA model as a constrained general diagnostic model: two variants of a model equivalency
- The generalized DINA model framework
- The sufficient and necessary condition for the identifiability and estimability of the DINA Model
- Training Products of Experts by Minimizing Contrastive Divergence
Cited in
(8)- Specification and identification issues in models involving a latent hierarchical structure
- Half-trek criterion for identifiability of latent variable models
- Identification with latent variables
- Hierarchical Identification With Pre-Processing
- A Joint MLE Approach to Large-Scale Structured Latent Attribute Analysis
- Latent conjunctive Bayesian network: unify attribute hierarchy and Bayesian network for cognitive diagnosis
- Designing learning intervention studies: identifiability of heterogeneous hidden Markov models
- Uncovering latent structures: a Bayesian approach to estimating Q-matrix and attribute hierarchies in cognitive diagnostic models
This page was built for publication: Identifiability of Hierarchical Latent Attribute Models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6069489)
