Review of test_outliers.py

[inglesp]

@CarolineMorton you asked me to review your test -- it all looks sensible and looks like it could pick up some bugs in the code. I'm a big fan of your approach of testing that a function works by implementing it a second way, and testing that both functions give the same result.

A couple of minor things:

• I'd like to see an example of the kind of thing that make_fake_data() would return -- you can work it out by reading the code, but that requires extra thinking
• In alternate_get_stats(), you never use a_num, b_num, or b_list

[CarolineMorton]

hi @inglesp

thanks for this. I will make the changes in alternative_get_stats(). When you say about make_fake_data... do you want the example data set printed to the terminal or put somewhere else for example the documentation.

[inglesp]

I'd thought the example data could be included in the docstring for make_fake_data().