search

ACCLAB / dabestr

Data Analysis with Bootstrap Estimation in R
https://acclab.github.io/dabestr
Apache License 2.0
211 stars 34 forks source link

Large dataset -- issue with bootstrap repetitions #123

Closed signebray closed 2 years ago

signebray commented 2 years ago

I get this error when running dabest on a large data set (>7k samples):

Error in bca.ci(boot.out, conf, index[1L], L = L, t = t.o, t0 = t0.o, : estimated adjustment 'a' is NA

A google search suggests this is because the number of bootstrap iterations needs to be increased.

josesho commented 2 years ago

Hi @signebray ,

You can adjust the bootstraps used with the reps argument:

your.tidy.data %>%
  dabest(xcol, ycol, 
         idx = c("Control1", "Group1"), 
         paired = FALSE) %>%
   mean_diff(reps = 10000)

The default value is 5000.

signebray commented 2 years ago

Thank you! I was looking at the dabest function options and did not see there but that makes sense.