A network estimation procedure for binary data, developed by Claudia van Borkulo and Sacha Epskamp. The method, named eLasso, fits Ising models using l1-regularized logistic regression and selects relevant variable relationships with the Extended Bayesian Information Criterion (EBIC). The resulting network represents variables as nodes and their relevant relationships as edges.
Use Cases
- Estimating interaction networks for binary variables based on the described eLasso method.
- Identifying relevant relationships between binary variables using the Extended Bayesian Information Criterion (EBIC).
- Applying l1-regularized logistic regression for model fitting on binary data as described.
Strengths
- Method is specifically designed to handle binary data, as stated in the description.
- Procedure combines two established techniques: l1-regularized logistic regression and model selection via EBIC.
Limitations
- Row count is unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
- Last update date is unknown; freshness unverified.
Provenance
- Source
- Claudia van Borkulo, Sacha Epskamp; with contributions from Alexander Robitzsch and Mihai Alexandru Constantin
- Collection Method
- Network estimation procedure (eLasso) for fitting Ising models.