ttScreening (version 1.8)

firth_screening: ttScreening with Firth bias-reduced logistic regression + multiple testing

Description

firth_screening: a two-stage approach combining ttScreening with Firth’s bias-reduced logistic regression. This method is designed to handle small sample sizes and rare outcomes by reducing bias in logistic regression coefficients.

Two-stage screening: (1) Pre-screen each predictor with bias-reduced logistic regression (brglm2). (2) For predictors passing pre-screen, run repeated train/test splits and count how often the test p-value < 0.05. Also report Bonferroni and FDR sets from the pre-screen p-values.

Dr. Abrar maintains and is the author of this functions, for questions contact him at [e-mail].

Usage

firth_screening(
  predictor_list,
  Outcome,
  iteration,
  train_frac = 0.67,
  alpha = 0.05,
  firth_count_threshold = 50,
  verbose = TRUE
)

Value

A list with:

• firth_tt: character vector of predictor IDs selected by repeated ttScreening (count >= firth_count_threshold)

• Bonferroni: character vector selected by Bonferroni-adjusted p < 0.05 (pre-screen)

• FDR: character vector selected by FDR-adjusted p < 0.05 (pre-screen)

• prescreen_p: named numeric vector of pre-screen p-values

• counts: named integer vector of ttScreening success counts

Arguments

predictor_list

Numeric matrix of dimension n x p (rows = subjects, cols = predictors). Column names are treated as predictor IDs; if missing, they will be generated.

Outcome

Integer or logical vector of length n with values in {0,1}.

iteration

Integer, number of train/test resampling iterations (e.g., 100).

train_frac

Numeric in (0,1); fraction of each class used for training (default 0.67).

alpha

Numeric in (0,1) p-value cutoff used inside splits (default 0.05).

firth_count_threshold

Integer, minimum number of "successes" to call a predictor selected by the ttScreening path (default 50).

verbose

Logical; if TRUE prints a short summary (default TRUE).

Examples

if (FALSE) {
res <- firth_screening(
  predictor_list, Outcome,
  iteration = 100, train_frac = 0.67, alpha = 0.05, firth_count_threshold = 50
)
length(res$firth_tt); head(res$firth_tt)
}