BEST is robust
The Bayesian model behind BEST assumes that the generative model for the data is a t-distribution; a more flexible distribution than the normal distribution as it assumes that data points might be outliers to some degree. This makes BEST's estimate of the mean difference robust to outliers in the data.
This exercise is part of the course
Fundamentals of Bayesian Data Analysis in R
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# The IQ of zombies given a regular diet and a brain based diet.
iq_brains <- c(44, 52, 42, 66, 53, 42, 55, 57, 56, 51)
iq_regular <- c(55, 44, 34, 18, 51, 40, 40, 49, 48, 46)
# Modify the data above and calculate the difference in means