Practice the following exercises to revise the concepts learned so far:
- Is multiple imputation amenable to parallel computation? Why, or why not?
- How is the way we call to.radians wasteful? Is there any way to refactor our code to use to.radians in a more efficient way?
- When I was gathering the data from Figure 12.2, I didn't check every sample size from 1 to the full data set; yet, I've obtained a smooth curve. What I did was test the performance of a handful of sample sizes from 100 to only 2,000. Then I used nls (non-linear least squares) to fit an equation of the form
(where n is the sample size) to the data points, and extrapolated with this equation after solving for x. What are some benefits and drawbacks of this approach? Do this on your own machine, if applicable. Do your performance curves match mine? - There is a thought among some scholars that there...
