Cobin and Micobin Regression Models for Continuous Proportional Data

A 5.4 MB collection of supplementary materials for a research article proposing improved regression models for continuous proportional data. The work, authored by Changwoo J. Lee, was last updated on May 18, 2026. It includes PDF, TXT, and DS_STORE files detailing the cobin and micobin models, which address robustness and computational issues in beta regression.

Use Cases

• Developing robust regression models for data bounded between 0 and 1 based on the described cobin and micobin distributions.
• Implementing Bayesian computation via Gibbs sampling for hierarchical models using the described Kolmogorov-Gamma data augmentation scheme.
• Analyzing ecological indices like the benthic macroinvertebrate multimetric index using watershed covariates as demonstrated in the article.
• Comparing the computational efficiency and outlier robustness of the proposed models against traditional beta regression.

Strengths

• Includes a 5.4 MB package of supplementary materials for reproducibility.
• The proposed models are described as having attractive robustness, computational, and flexibility properties for handling boundary values (0 or 1).
• License is CC-BY-4.0, permitting broad reuse with attribution.

Limitations

• The dataset consists of supplementary article materials (PDF, TXT); the primary data for the benthic macroinvertebrate index analysis is likely not included.
• Row count and column-level documentation for any underlying data are unknown.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Supplementary materials for a research article.

Freshness

Last updated 2026-05-18 13:59:11; freshness should be verified.

Geography

Analysis example uses U.S. lakes, but geographic scope of the models is general.