Exercise

# 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.

Instructions

**100 XP**

- Create an array
`f_shift`

, by shifting`f`

such that its mean is zero. You can use the variable`f_mean`

computed in previous exercises. - Draw 100,000 bootstrap replicates of the mean of the
`f_shift`

. - Compute and print the p-value.