R/DMRnet4glm_help.R

In DMRnet: Delete or Merge Regressors Algorithms for Linear and Logistic Model Selection and High-Dimensional Data

DMRnet4glm_help <- function(S, X, y, fl, clust.method, lam){
    if (sum(S) == 0) {
       m <- stats::glm.fit(as.matrix(rep(1, length(y))), y, family = stats::binomial())
       zb = exp(m$coef*rep(1, length(y)))
       pix = zb/(zb + 1)
       loglik = sum(log(pix)[y == 1]) + sum(log(1-pix)[y == 0])
       b = c(m$coef, rep(0,sum(fl-1)))    #SzN:fixing that line in accordance with 4lm version and to avoid crashes in adult dataset in GLAMER
       return(list(b = b, loglik = loglik))
    }
    Xn <- X[, S==1, drop = FALSE]
    mfin <- DMR4glm_help(Xn, y, clust.method = clust.method, lam = lam)
    return(mfin)
}