Skip to contents

First Significant Cutpoint per Feature

Source: R/batch_cutpoint_survdiff.R
first_significant_cutpoint.Rd

From the result of [batch_cutpoint_survdiff()], picks for each feature and each `alpha` the first cutpoint with `p < alpha`, scanning from the lowest cutpoint when the univariate Cox coefficient is positive and from the highest otherwise (the most inclusive split on the risk side).

Usage

first_significant_cutpoint(x, alpha = c(0.1, 0.05, 0.001))

Arguments

x

Result of [batch_cutpoint_survdiff()].

alpha

Significance levels (default `c(0.1, 0.05, 0.001)`).

Value

A tibble with one row per feature and `alpha` that has a significant cutpoint (the columns of `x` plus `alpha`).

See also

[batch_cutpoint_survdiff()]

Examples

set.seed(1)
d <- data.frame(time = rexp(80), status = rbinom(80, 1, 0.7), x = rnorm(80))
first_significant_cutpoint(batch_cutpoint_survdiff(d, "time", "status", "x"), alpha = 0.5)
#> # A tibble: 1 × 10
#>   feature cutpoint n_low n_high chisq     p p_type   cox_coef cox_p alpha
#>   <chr>      <dbl> <int>  <int> <dbl> <dbl> <ord>       <dbl> <dbl> <dbl>
#> 1 x          -1.46     1     79 0.627 0.428 p >= 0.1  0.00479 0.975   0.5