DSFM (version 1.0.1)

DPPC: Distributed Probabilistic Principal Component Analysis

Description

Performs distributed probabilistic principal component analysis (PPC) on a numeric dataset split across multiple nodes. Estimates loading matrices, residual variances, and covariance matrices for each node using a probabilistic approach.

Usage

DPPC(data, m, n1, K)

Value

A list with the following components:

Apro

List of estimated loading matrices for each node.

Dpro

List of diagonal residual variance matrices for each node.

Sigmahatpro

List of covariance matrices for each node.

Arguments

data

A numeric matrix containing the total dataset.

m

An integer specifying the number of principal components.

n1

An integer specifying the length of each data subset.

K

An integer specifying the number of nodes.

Examples

set.seed(123)
data <- matrix(rnorm(500), nrow = 100, ncol = 5)
DPPC(data = data, m = 3, n1 = 20, K = 5)