Generalized Grade-of-Membership Model Data for Political Surveys and Genetics

A 1.4 MB dataset uploaded by Ling Chen on figshare in May 2026, associated with a research paper on mixed membership models. The data likely contains multivariate categorical responses, such as from political surveys or population genetics studies, used to demonstrate a novel spectral estimation method. The work addresses estimation challenges for high-dimensional polytomous data with locally dependent noise.

Use Cases

• Testing spectral estimation methods for grade-of-membership models based on the described three-way tensor formulation.
• Analyzing survey response patterns in political science based on the mention of political survey applications.
• Modeling population structure in genetics data based on the described use case for population genetics.
• Benchmarking high-dimensional categorical data analysis techniques against the proposed generalized-GoM framework.

Strengths

• Dataset is 1.4 MB in size, indicating a manageable download and storage footprint.
• Associated with a research paper that establishes finite-sample error bounds for the proposed method, suggesting a theoretical foundation.
• The description mentions applications to political surveys, population genetics, and single-cell sequencing, indicating cross-domain relevance.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• The primary files are PDF and ZIP formats; the actual data structure within the archive requires inspection.

Provenance

Source

Ling Chen via figshare.

Collection Method

Likely generated or compiled for research on grade-of-membership model estimation.

Freshness

Last updated 2026-05-18 13:46:02; freshness should be verified.