1. Write a function (based on the above) that will generate, for a combinations of 'sample size' and 'error', a measure of the quality of the parameter estimate. The measure you can use is the t-value, reported by the summary function. Use a value of B1 = 3 and B0 = 0.
2. Use that function to produce a data frame, with combinations of each of the two variables as 'factors' and the quality of the estimate as the response. There are several ways of solving this problem, but one is to use expand_df to create a 'null' data frame, then fill it with the response.
3. Construct a single plot that shows how the value of the response varies with the two variables.
4. Do the same thing as above, but use a single parameter estimate of B1 = 0 (in other words, no effect). (Set B0 = 0). Use a sample size of 100. Instead of using the 't' value as your 'quality' estimate, have the function output the 'p-value'.
4a. For a given sample size and error, what does the distribution of p-values look like? 4b. What proportion of p-values are 4c. How does that change when the amount of error changes.