Exercise

# How should we do a hypothesis test on differences of the b-value?

We wish to test the hypothesis that the *b*-value in Oklahoma from 1980 through 2009 is the same as that from 2010 through mid-2017. Which of the first five statements is false? If none of them are false, select the last choice.

Instructions

**50 XP**

##### Possible Answers

- You should only include earthquakes that have magnitudes above the completeness threshold. A value of 3 is reasonable.
- You should perform a permutation test because asserting a null hypothesis that the
*b*-values are the same implicitly assumes that the magnitudes are identically distributed, specifically Exponentially, by the Gutenberg-Richter Law. - A reasonable test statistic is the difference between the mean post-2010 magnitude and the mean pre-2010 magnitude.
- You do not need to worry about the fact that there were far fewer earthquakes before 2010 than there were after. That is to say, there are fewer earthquakes before 2010, but sufficiently many to do a permutation test.
- You do not need to worry about the fact that the two time intervals are of different length.
- None of the above statements are false.