cv.gspca: CV for generalized supervised PCA

In andland/genSupPCA: Supervised Dimensionality Reduction for Exponential Family Data

Description

Run cross validation on dimension and alpha for generalized supervised PCA

Usage

cv.gspca(x, y, ks, alphas = 10^seq(-3, 5, by = 0.5), family_x, family_y,
  folds = 5, quiet = TRUE, ...)

Arguments

x	covariate matrix
y	response vector
ks	the different dimensions k to try
alphas	the different approximations to the saturated model m to try
family_x	exponential family distribution of covariates
family_y	exponential family distribution of response
folds	if folds is a scalar, then it is the number of folds. If it is a vector, it should be the same length as the number of rows in x
quiet	logical; whether the function should display progress
...	Additional arguments passed to logisticPCA

Value

A matrix of the CV deviance for the response with k in rows and alpha in columns

Examples

# construct a low rank matrix in the logit scale
rows = 100
cols = 10
set.seed(1)
mat_logit = outer(rnorm(rows), rnorm(cols))

# generate a binary matrix
mat = (matrix(runif(rows * cols), rows, cols) <= inv.logit.mat(mat_logit)) * 1.0

## Not run: 
negloglikes = cv.gspca(mat, ks = 1:9)
plot(negloglikes)

## End(Not run)

andland/genSupPCA documentation built on May 30, 2019, 11:43 a.m.