Elliptical Factor Models
The EFM function is to generate Elliptical Factor Models data.
Calculate errors for elliptical distributions
SOPC Estimation Function for Elliptical Factor Model
Apply the FanPC method to the Elliptical Factor Model
Incremental Perturbation PCA
Apply the PC method to the Elliptical Factor Model
PPC: Principal Projection Components
Apply the PPC method to the Elliptical Factor Model
Random Perturbation Robust PCA
The elliptical factor model, as an extension of the traditional factor model, effectively overcomes the limitations of the traditional model when dealing with heavy-tailed characteristic data. This package implements sparse principal component methods (SPC) and bi-sparse online principal component estimation (SPOC) for parameter estimation. Includes functionality for calculating mean squared error, relative error, and loading matrix sparsity.The philosophy of the package is described in Guo G. (2023) <doi:10.1007/s00180-022-01270-z>.