ppcaMLE: MLEs for the PPPCA model

In WeiAkaneDeng/SPAC2: Statistical and Probabilistic Algorithms for Classification and Clustering

Description

The function returns the MLEs for the PPCA model at a given choice of dimension. This function is adapted from the one made available by Mark Clark on github with the link of the original code under “See also”.

Usage

ppcaMLE(x, nComp = 2, tol = 1e-06, maxits = 100)

Arguments

x	a data matrix with the number of rows to be reduced; only complete columns are used.
nComp	an integer specifying the number of principal components or effective dimension retained.
tol	a tolerance level for the EM algorithm to terminate computations.
maxits	the maximum number of iterations for EM to converge to the MLEs.

Value

a list of MLEs: the first item of the list is the loading matrix W, and the second item of the list is the error variance σ^{2} or sigma2.

References

Tipping, M. E., and Bishop, C. M. (1999). Probabilistic principal component analysis. **Journal of the Royal Statistical Society: Series B (Statistical Methodology)**, *61*(3), 611-622. <doi:10.1111/1467-9868.00196>

See Also

https://github.com/m-clark/Miscellaneous-R-Code/blob/master/ModelFitting/EM%20Examples/EM%20algorithm%20for%20ppca.R

Examples

## Not run: 
library(MASS)
X <- mvrnorm(1000, mu = rep(0,10), Sigma = diag(1,10))
ppcaMLE(x = t(X), nComp = 5)

## End(Not run)

WeiAkaneDeng/SPAC2 documentation built on Jan. 15, 2022, 5:01 a.m.