Logarithmic Models

1 Find the definition of g(r). What does this function do?

2 Find the Contract for build-column on the Contracts Page.

3 At the end of the program, you’ll find this code:

4 Click "Run", and evaluate countries-transformed in the Interactions Area on the right to test it out!

a. What did you get back?

b. What is different about this Table?

c. Where did the column on the right come from?

d. What does that line of code at the end of the program do?

5 Use this new table to make an lr-plot comparing log(pc-gdp) and median-lifespan with country as the label.

6 Record the regression line here: 𝑦 = slope 𝑥 + vertical shift 𝑅²:

7 Use the values produced by lr-plot to complete the model: 𝑙o𝑔𝑎𝑟𝑖𝑡ℎ𝑚𝑖𝑐(𝑥) = log coefficient 𝑙o𝑔₁₀(𝑥) + vertical shift

8 Let’s interpret this model:

A 10x increase in per-capita-GDP is associated with median-lifespan increasing by .

9 Rewrite your model as a pyret definition: fun logarithmic(x): end

10 Add the definition to your starter file and click "Run" to load it.

11 Use fit-model to calculate the value of 𝑅²:

12 Let’s interpert this 𝑅² value:

Roughly percent of the variation in is explained by the variation in .

★ Are there other relationships you can think of, which might be logarithmic?