Scalable Skewed Bayesian Inference for Latent Gaussian Models Using INLA and VB

A 201.9 KB dataset by Shourya Dutta, last updated April 21, 2026, illustrating a skewed extension of the Integrated Nested Laplace Approximation (INLA) method for Bayesian inference. The approach is formulated for scalability and is demonstrated using simulated cases and applications to a rare disease and a large diabetes dataset. The dataset is shared under a CC-BY-4.0 license on figshare.

Use Cases

• Benchmarking scalable Bayesian inference methods based on the described skewed Laplace and variational framework.
• Analyzing class imbalance in binary regression problems, as mentioned in the rare disease application.
• Testing inference performance on large datasets, referenced by the diabetes dataset application.
• Studying the effect of heavy-tailed likelihoods on posterior approximations using the proposed methodology.

Strengths

• Methodology is explicitly formulated for scalability in model complexity and data size.
• Includes applications to real-world problems like a rare disease and a large diabetes dataset.
• Shared under a permissive CC-BY-4.0 license for open use.

Limitations

• Dataset size is 201.9 KB, indicating a very limited scope, likely containing illustrative results rather than extensive raw data.
• Row count and column-level documentation are unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality and structure require manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely contains results from simulated cases and applied analyses described in the associated methodological paper.

Time Range

null

Freshness

Last updated 2026-04-21 19:16:54; freshness should be verified.

Geography

null