fdapace: Functional Data Analysis Package for Sparse and Dense Trajectories

fdapace is an R package implementing Functional Data Analysis and Empirical Dynamics methods, with Functional Principal Component Analysis (FPCA) via the PACE algorithm as its core. The package is designed for analyzing sparsely or densely sampled random trajectories and time courses, providing fitted continuous trajectories with confidence bands for sparse designs. Updates were supported by NIH Echo and NSF grants DMS-1712864 and DMS-2014626.

Use Cases

• Perform Functional Principal Component Analysis (FPCA) on longitudinal data based on the PACE algorithm described.
• Model sparse longitudinal observations as continuous trajectories with confidence bands based on the package's capability for sparse designs.
• Analyze empirical dynamics and functional derivatives based on the package's stated support for functional data and derivatives.

Strengths

• Implements the Principal Analysis by Conditional Estimation (PACE) algorithm, a key FDA technique.
• Supports both dense (functional) and sparse (longitudinal) sampling designs.
• Provides fitted continuous trajectories with confidence bands for subjects with very few observations.

Limitations

• Row count and data scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

Software package release and documentation.