cv.lme1en: cv.lme1en

In oyvble/lme1en: Elastic Net With Random Intercept

Description

Performing crossvalidation of lme1en (extension of inputs)

Usage

cv.lme1en(y, X, batch, rho = 0, lambda = 1, alpha = 0.5,
  beta = NULL, glmnetPenalty = TRUE, glmnetWarmup = TRUE,
  maxit = 10000, toler = 1e-05, verbose = FALSE, errorFun = NULL,
  Kfold = 10, CViter = 10, parallel = TRUE, seed = NULL)

Arguments

y	response vector. Must be normalized before input.
X	Design matrix belonging to fixed effects coefficients (beta). Must be normalized before input.
batch	factor with batch effect names (vector for each observations)
rho	proportion of variation explained by batch effect
lambda	numeric, penalty levels for fixed effects betas (component 1)
alpha	numeric, penalty levels for fixed effects betas (balancing LASSO/RIDGE erros)
beta	numeric, initial values of the beta coefficients (using glmnet or marginal estimates if not provided)
glmnetPenalty	boolean, whether to use the original peanalty (FALSE) or the glmnet penalty (TRUE)
glmnetWarmup	boolean, whether to use glmnet beta-estimates as warmup (if not marginals are used)
maxit	maximum number of iterations (i.e. forloops) in the coordinate decent algorithm
toler	tolerance level of beta changes for each iterations (similar to 'thresh' in glmnet)
verbose	boolean, show progress Default: FALSE
errorFun	function, The user can provide own error function to perform cross validation with (Default is MSE if function not provided). The function must be able to take equally long two vectors and return a number
Kfold	The train-test fold split: The training proportion becomes '(Kfold-1)/Kfold'
CViter	Number of iterations of the cross-validation
parallel	Boolean of whether parallelization in CV process should be applied (utilizing all CPUs)
seed	The user can set seed for reproducibility if wanted

Details

The function performs (possibly parallelized) K-fold cross validation for given alpha,lambda,rho The function assume that the response y and design matrix X are centralized (no intercept returned from function)

Value

matrix with error values (test and train) for each iterations

Examples

## Not run:  
dat = genData(seed=1)
CVerrors = cv.lme1en(y=dat$y,X=dat$X,batch=dat$batch, rho=0.3,verbose=TRUE,seed=1)

## End(Not run)

oyvble/lme1en documentation built on April 30, 2020, 2:41 p.m.