Hypothesis test: Does lane assignment affect performance?
Perform a bootstrap hypothesis test of the null hypothesis that the mean fractional improvement going from low-numbered lanes to high-numbered lanes is zero. Take the fractional improvement as your test statistic, and "at least as extreme as" to mean that the test statistic under the null hypothesis is greater than or equal to what was observed.
This exercise is part of the course
Case Studies in Statistical Thinking
Exercise instructions
- Create an array
f_shift, by shiftingfsuch that its mean is zero. You can use the variablef_meancomputed in previous exercises. - Draw 100,000 bootstrap replicates of the mean of the
f_shift. - Compute and print the p-value.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Shift f: f_shift
f_shift = ____ - ____
# Draw 100,000 bootstrap replicates of the mean: bs_reps
bs_reps = ____
# Compute and report the p-value
p_val = ____(____ >= ____) / 100000
print('p =', p_val)