The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

Tutorial: Obtain an overall p-value for a factor variable

Emily C. Zabor

Last updated: January 18, 2024

Context

After using eh_test_subtype() to obtain a model fit, if factor variables are involved in the analysis it will be of interest to obtain overall p-values testing for differences across subtypes across all levels of the factor variable.

The posthoc_factor_test() function allows for post-hoc testing of a factor variable.

Example

# Load needed packages
library(riskclustr)
library(dplyr)
# create a new example dataset that contains a factor variable
factor_data <- 
  subtype_data %>%
  mutate(
    x4 = cut(
      x1,
      breaks = c(-3.4, -0.4, 0.3, 1.1, 3.8),
      include.lowest = T,
      labels = c("1st quart",
                 "2nd quart",
                 "3rd quart",
                 "4th quart")
      )
    )
# Fit the model using x4 in place of x1
mod1 <- eh_test_subtype(
  label = "subtype",
  M = 4,
  factors = list("x4", "x2", "x3"),
  data = factor_data,
  digits = 2
)

After we have the model fit, we can obtain the p-value testing all levels of x4 simulaneously.

mypval <- posthoc_factor_test(
  fit = mod1, 
  factor = "x4", 
  nlevels = 4
  )

The function returns both a formatted and unformatted p-value. The formatted p-value can be accessed as pval:

mypval$pval
#>      [,1]   
#> [1,] "<.001"

The unformatted p-value can be accessed as pval_raw:

mypval$pval_raw
#>      [,1]
#> [1,]    0

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.