subgrp_perf {BioPred}R Documentation

Subgroup Performance Evaluation for Prognostic Cases

Description

This function evaluates subgroup performance based on different types of response variables.

Usage

subgrp_perf(
  yvar,
  censorvar = NULL,
  grpvar,
  grpname,
  xvars.adj = NULL,
  data,
  type,
  yvar.display = yvar,
  grpvar.display = grpvar
)

Arguments

yvar

The response variable name.

censorvar

(Optional) The censoring variable name (0-censored, 1-event).

grpvar

The subgroup variable name.

grpname

A vector of ordered subgroup names (values in the column of grpvar).

xvars.adj

(Optional) Other covariates to adjust when evaluating the performance.

data

The dataset containing the variables.

type

The type of response variable: "c" for continuous, "s" for survival, and "b" for binary.

yvar.display

Display name of the response variable.

grpvar.display

Display name of the group variable.

Value

A list containing subgroup performance results including logrank p-value, median and mean survival, Cox model p-value, ANOVA p-value, and more based on the specified response variable type.

Examples

# Load a sample dataset
data <- data.frame(
  survival_time = rexp(100, rate = 0.1),  # survival time
  status = sample(c(0, 1), 100, replace = TRUE),  # censoring status
  group = sample(c("Low", "Medium", "High"), 100, replace = TRUE),  # subgroup variable
  covariate = rnorm(100, mean = 50, sd = 10)  # an additional covariate
)

# Perform subgroup performance evaluation for survival analysis
subgrp_perf(
  yvar = "survival_time",
  censorvar = "status",
  grpvar = "group",
  grpname = c("Low", "Medium", "High"),
  data = data,
  type = "s",
  yvar.display = "Survival Time",
  grpvar.display = "Risk Group"
)

# Perform subgroup performance evaluation for continuous outcome
data$continuous_outcome <- rnorm(100, mean = 10, sd = 5)
subgrp_perf(
  yvar = "continuous_outcome",
  grpvar = "group",
  grpname = c("Low", "Medium", "High"),
  data = data,
  type = "c",
  yvar.display = "Continuous Outcome",
  grpvar.display = "Risk Group"
)

# Perform subgroup performance evaluation for binary outcome
data$binary_outcome <- sample(c(0, 1), 100, replace = TRUE)
subgrp_perf(
  yvar = "binary_outcome",
  grpvar = "group",
  grpname = c("Low", "Medium", "High"),
  data = data,
  type = "b",
  yvar.display = "Binary Outcome",
  grpvar.display = "Risk Group"
)

[Package BioPred version 1.0.2 Index]