Preparations for lecture on Hypothesis Testing

As a preparation for the Hypothesis Testing convocation, make yourself acquainted with the material below, and subsequently complete any tasks assigned to you in canvas.

1. Watch the Video Lecture on Hypothesis Testing, and its slides
2. Read the chapter 8.3 and 9.1-9.5 in Downey on hypothesis testing. The associated code is available here.
   • The book gives a different method for how to derive p values from samples than most other books. They derive p values from simulations, rather than to look them up from formulas or tables. If you prefer a traditional explanation, try e.g. wikipedia’s entry on t tests.
   • Unfortunately, Downey does not give a distinct definition of a p value. The correct definition is “the probability of obtaining test results at least as extreme as the results actually observed, under the assumption that the null hypothesis is correct.”, however, in the glosary 9.13 an other definitions is given. Try to convince yourself that you understand why the glosaries definitions is problematic. See for instance this wikipedia entry.
3. Investigate the jupyter notebook on Differential expression anlaysis of the TCGA breast cancer set. If you want some study questions we provide some here
4. Read previous years questions and answers on the material.

Additional/Alternative material

1. StatQuest on Hypothesis testing and p values
2. Khan Academy on Hypothesis testing
3. Wikipedia entries for p value, p value falacy, Volcano plots and permutation tests.
4. Stackexchange comments on Fisher vs. Neyman & Pearson framework
5. XKCD on p-hacking