Throughout our discussion of regression analysis, we used the Regression command to obtain the parameter estimates that minimize the sum of squared estimation errors. Suppose that we want to obtain parameter estimates that minimize the absolute value of the maximum estimation error, or:

a. Use Solver to obtain the parameter estimates for a simple linear regression function that minimizes the absolute value of the maximum estimation error for the data in question 9.
b. What advantages, if any, do you see in using this alternate objective to solve a regression problem?
c. What disadvantages, if any, do you see in using this alternate objective to solve a regressionproblem?

Transcribed Image Text:

Y-Y