| Title: | Unsmoothed and Smoothed Penalized PCA using Nesterov Smoothing |
| Version: | 1.0.0 |
| Maintainer: | Rebecca Hurwitz <rebeccahurwitz@hsph.harvard.edu> |
| Description: | We provide functionality to implement penalized PCA with an option to smooth the objective function using Nesterov smoothing. Two functions are available to compute a user-specified number of eigenvectors. The function unsmoothed_penalized_EV() computes a penalized PCA without smoothing and has three parameters (the input matrix, the Lasso penalty, and the number of desired eigenvectors). The function smoothed_penalized_EV() computes a smoothed penalized PCA using the same parameters and additionally requires the specification of a smoothing parameter. Both functions return a matrix having the desired eigenvectors as columns. |
| License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.2.3 |
| Imports: | stats |
| NeedsCompilation: | no |
| Packaged: | 2023-08-16 18:52:44 UTC; rebeccahurwitz |
| Author: | Rebecca Hurwitz [aut, cre], Georg Hahn [ctb] |
| Repository: | CRAN |
| Date/Publication: | 2023-08-17 14:42:33 UTC |
smoothed_penalized_EV
Description
This function takes a matrix (m), a lambda value (lambda), the number of desired eigenvectors (k), and a mu value (mu) as input. It then computes eigenvectors 1 to k, penalized by the supplied lambda and smoothed by the Nesterov smoothing function.
Usage
smoothed_penalized_EV(m, lambda, k, mu)
Arguments
m |
A matrix generated from a large dataset. |
lambda |
A numeric vector of lambda values to use for the penalty. |
k |
The number of eigenvectors we consider in the analysis. |
mu |
A number assigned to mu; we are typically using 0.1. |
Value
Returns smoothed eigenvectors 1 to k for the specified lambda value.
Examples
# Generate a small matrix for testing
m <- matrix(rnorm(100), nrow = 10)
# Call function (using matrix, lambda, mu, and k)
smoothed_penalized_EV(
m = m,
lambda = 1,
k = 2,
mu = 0.1
)
unsmoothed_penalized_EV
Description
This function takes a matrix (m), a lambda value (lambda), and the number of desired eigenvectors (k) as input. It then computes eigenvectors 1 to k, penalized by the supplied lambda.
Usage
unsmoothed_penalized_EV(m, lambda, k)
Arguments
m |
A matrix generated from a large dataset. |
lambda |
A numeric vector of lambda values to use for the penalty. |
k |
The number of eigenvectors we consider in the analysis. |
Value
Returns eigenvectors 1 to k for the specified lambda value.
Examples
# Generate a small matrix for testing
m <- matrix(rnorm(100), nrow = 10)
# Call function (using matrix, lambda, and k)
unsmoothed_penalized_EV(
m = m,
lambda = 1,
k = 2
)