Assigned Readings

Links are either to open resources, or canvas (only accessible to registered students).

Week 0

Pre-lecture reading:

Read the article: Putin’s Peaks: Russian Election Data Revisited. Kobak, Dmitry, Sergey Shpilkin, and Maxim S. Pshenichnikov. n.d. Significance 15 (3): 8–9

Be prepared to answer the following questions in class:

1. What feature of Figure 1 do the authors argue is evidence for election fraud?

2. For a particular year, say 2018, how do the authors measure the statistical significance of the feature in Figure 1? What statistic do they calculate? What do they compare that statistic to?

Week 1

Pre-lecture reading:

• Section 3.1 Pseudorandom Number Generation in Simulation (1 page)
• Section 4.1 The Inverse Transform Method in Simulation up to and including Example 4a (~ 2 pages)
• Section 5.1 The Inverse Transform Algorithm in Simulation up to and including Example 5b (~ 2 pages)

Be prepared to answer:

• Why are random numbers generated by a computer called pseudo-random numbers?
• How would you use a Uniform(0, 1) random variable to simulate a Bernoulli($\pi$) random variable?
• How would you use a Uniform(0, 1) random variable to simulate a Exponential($\lambda$) random variable?

Week 2

Pre-lecture reading:

Functions in R4DS 19.1 through and including 19.3 (If you are on a roll keep reading…)

Be prepared to answer:

• What are the three key steps to writing a function?
• Do exercise 19.2.1 #3

Week 3

Pre-lecture reading:

Wilson G, Bryan J, Cranston K, Kitzes J, Nederbragt L, et al. (2017) Good enough practices in scientific computing. PLOS Computational Biology 13(6): e1005510. https://doi.org/10.1371/journal.pcbi.1005510

a minimum set of tools and techniques that we believe every researcher can and should consider adopting.

Read the Author summary, Overview, Introduction and Software sections.

You may also want to review some things we touched on in class about functions in R:

• Provide defaults for detail arguments not data arguments
• Choosing names for arguments is important too
• … dot-dot-dot
• Functions can find variables in the global environment

Week 4

Section 1.2 Introduction (page 8) up to and including page 10, in Bootstrap Methods: A Guide for Practitioners and Researchers by Michael R. Chernick

Be prepared to answer the following questions:

• Usually when thinking about constructing an estimate for a paramater, we make some assumptions about the population distribution. What does the bootstrap replace the population distribution with? Why is this a good replacement?

• Why do we need a Monte Carlo approach?

If you’ve seen the bootstrap before, you might continue reading to the end on the Introduction, paying attention to the reasons the bootstrap might not work.