LFM: An R package for laplace factor model

The Laplace Factor Model (LFM) is a valuable mathematical tool used in statistics, machine learning, and data analysis. It uses the Laplace distribution to capture data sparsity and uncertainty, effectively handling complex, large-scale data. The proposed R package, called LFM, has the capability to...

Full description

Saved in:
Bibliographic Details
Main Authors: Siqi Liu, Guangbao Guo
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:SoftwareX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352711025001001
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Laplace Factor Model (LFM) is a valuable mathematical tool used in statistics, machine learning, and data analysis. It uses the Laplace distribution to capture data sparsity and uncertainty, effectively handling complex, large-scale data. The proposed R package, called LFM, has the capability to construct factor models based on the Laplace distribution, and it allows for customized model building by flexibly adjusting the parameters of the Laplace distribution. Additionally, the LFM package integrates various techniques including Sparse Online Principal Component (SOPC), Incremental Principal Component (IPC), Projection Principal Component (PPC), Stochastic Approximate Principal Component (SAPC), Sparse Principal Component (SPC), and other PC methods and the Farm Test method. By evaluating indicators such as the accuracy of parameter estimation, mean square error, and sparsity, this study verifies the effectiveness and practicality of these methods in the Laplace Factor Model.
ISSN:2352-7110