“SFM”: An R package for Skew Factor Models

Abstract

The goal of the Skew Factor Model (SFM) package is to analyze datasets where factor loading matrices exhibit skewed distributions, providing efficient estimation methods for SFM. The proposed R package, called SFM, is specifically designed for estimating skewed factor structures and handling high-dimensional data with non-Gaussian distributions.It achieves this by implementing multiple principal component methods, including Sparse Online Principal Component, Incremental Principal Component, Projected Principal Component, Stochastic Approximation Principal Component, Sparse Principal Component and other Principal Component methods. Additionally, SFM package provides evaluation metrics such as mean squared error, relative error, and sparsity of the loading matrix, ensuring robust parameter estimation.