Robust Functional PCA for Relative Data with Simulation and Application Files

Jeremy Oguamalam's research dataset from 2026 introduces a robust method for functional principal component analysis (FPCA) on relative data. The 75.7 MB collection includes files for simulations and real-world applications validating the proposed RRPCA method. Data is shared under a CC-BY-4.0 license on the figshare platform.

Use Cases

• Developing robust statistical models for relative functional data based on the described Bayes space framework.
• Testing covariance estimation methods in the presence of outliers as referenced in the simulation validation.
• Comparing traditional FPCA performance against the proposed robust relative principal component analysis (RRPCA) method.
• Applying the regularized Mahalanobis distance concept for anomaly detection in constrained functional data.

Strengths

• Includes multiple file formats (PDF, RData, CSV, R, etc.) supporting reproducibility and analysis.
• Dataset size is 75.7 MB, indicating it contains substantial supporting material for the research.
• License is clearly defined as CC-BY-4.0, permitting open reuse with attribution.

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.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely generated through statistical simulations and real-world applications as part of methodological research.

Time Range

null

Freshness

Last updated 2026-04-30 13:33:45; freshness should be verified.

Geography

null