glmbystrata: Fit Linear or Logistic Regression GLM Model Within Strata

In rmarceauwest/fiveSTAR: Run 5-STAR Algorithm

Fit Linear or Logistic Regression GLM Model Within Strata

Description

Fits a linear or logistic regression GLM model within each formed strata, providing estimate of regression coefficient or log odds ratio (EXPERIMENTAL!)

Usage

glmbystrata(yy, arm, family, treeStrata, termNodes = NULL,
  cilevel = 0.05, verbose = 0, alternative = "two.sided")

Arguments

yy	Continuous response or case control status response vector
arm	Treatment indicator, 1 = test treatment, 0 = control
family	Family for glm; options: "binomial" or "gaussian"
treeStrata	Vector of strata membership, as defined by ctree strata
termNodes	Vector of names for each tree strata formed (e.g., number of strata or covariate definition)
cilevel	Confidence level alpha for overall result and confidence intervals
verbose	Numeric variable indicating amount of information to print to the terminal (0 = nothing, 1 = notes only, 2 = notes and intermediate output)

Value

• fitsummary: summary of GLM fits within each strata

• table: Summary of amalgamated regression coefficient or odds ratio estimate, variance, p-value, and cilevelx100% confidence interval assuming sample size weights

• weights: Sample size weights used to construct estimate

rmarceauwest/fiveSTAR documentation built on June 30, 2023, 7:38 a.m.