Est.ALASSO.GLMNET.TANGXIE: A wrapper for divide-and-conquer adaptive lasso proposed by...

In michaelyanwang/divideconquer: Divide-and-conquer adaptive lasso using least square approximation

Description

Est.ALASSO.GLMNET.TANGXIE fits adaptive lasso based on cv.glmnet. The best lambda (penalizing factor) is chosen by 10-fold cross-validation.

Usage

Est.ALASSO.GLMNET.TANGXIE(dat.list, K, BIC.factor = 0.1, fam0 = "binomial",
  lambda.grid, mvpct = 0.5, modBIC = T, adaptive = T, onestep = F)

Arguments

dat.list	a list of matrices. Each element of the list is a sub-dataset. In each sub-dataset, for non-survival outcome, first column = response, rest = design matrix. In each sub-dataset, for continuous-time survival outcome, first column = time, second column = delta (0/1), rest = design matrix.
K	Number of sub-datasets in dat.list
BIC.factor	factor in modified BIC, BIC = -2 loglikelihood + df * N^BIC.factor
fam0	family of the response, taking "binomial", "Poisson", "Cox"
lambda.grid	the grid of lambda to put into glmnet
mvpct	majority voting percentage used for Chen and Xie (2014)
modBIC	if the program also finds the lamda that minimizes the modified BIC. The default is TRUE.
adaptive	if adaptive lasso is used. The default is TRUE
onestep	if one-step estimator should be used as the initial estimator for Cox fit
w.b	w.b used to penalize adaptive lasso. If null, a glm/Cox model will be fitted and 1/abs(coefficients) will be used as w.b

Value

a list containing two arguments: bhat.cv adaptive lasso estimator using 10-fold cross-validation; lambda.cv is the optimal lambda chosen by cross-validation.

Author(s)

Yan Wang, Tianxi Cai

References

Chen, Xueying, and Min-ge Xie. "A split-and-conquer approach for analysis of extraordinarily large data." Statistica Sinica (2014): 1655-1684.

Tang, Lu, Ling Zhou, and Peter X-K. Song. "Method of Divide-and-Combine in Regularised Generalised Linear Models for Big Data." arXiv preprint arXiv:1611.06208 (2016).

Examples

Est.ALASSO.GLMNET.TANGXIE(dat.list,K,BIC.factor=0.1,fam0="binomial",lambda.grid,mvpct = 0.5)

michaelyanwang/divideconquer documentation built on Aug. 16, 2019, 10:11 a.m.