Limiting Laws for Spiked Eigenvalues in High-Dimensional Models

A dataset supporting a study on limiting laws and estimation criteria for extreme eigenvalues in high-dimensional spiked covariance models. The dataset likely contains tabular data from simulation studies and three real-world applications used to validate the proposed estimation methods. The research allows both the dimension p and the number of spikes k to diverge, establishing results under Gaussian and general distributions.

Use Cases

• Testing eigenvalue estimation criteria based on derived limiting distributions.
• Simulating spiked covariance models with diverging dimension p and number of spikes k.
• Applying consistent estimation methods to real-world datasets to determine the number of spikes.
• Comparing estimation performance under Gaussian versus general distributions with finite fourth moments.

Strengths

• Dataset is associated with a detailed methodological paper published on figshare.
• The underlying research establishes results under a general spiked covariance model, a condition not uniformly assumed in prior literature.

Limitations

• Column names and row counts are not provided in the metadata, limiting understanding of the data structure.
• The dataset's real-world applications are not described, making it difficult to assess its direct utility.

Provenance

Source

Jianwei Hu

Collection Method

Likely generated from simulation studies and collected from three real-world applications.

Freshness

2026-03-20