a. When the model Y
0 1x1 2x2 3 x2 1 4 x2 2

5 x1x2 ! is fit to the hops data of Exercise 49, the estimate of 5 is ˆ

5
.557 with estimated standard deviation sˆ

5
.94. Test H0: 5
0 versus Ha: 5 0.

b. Each t-ratio ˆ

i

/sˆ

i (i
1, 2, 3, 4, 5) for the model of part

(a) is less than 2 in absolute value, yet R2
.861 for this model. Would it be correct to drop each term from the model because of its small t-ratio? Explain.

c. Using R2
.861 for the model of part (a), test H0: 3

4
5
0 (which says that all second-order terms can be deleted).