For a data set for 100 adults on y = height, x1 =

length of left leg, and x2 = length of right leg, the model E(y) = α + β1x1 + β2x2 is fitted. Neither H0: β1 = 0 nor H0: β2 = 0 has a P-value below 0.05.

(a) Does this imply that length of leg is not a good predictor of height?Why?

(b) Does this imply that H0: β1 = β2 = 0 would not have a P-value below 0.05?Why?

(c) Suppose ryx1

= 0.901, ryx2

= 0.902, and rx1x2

= 0.999.

Using forward selection and the potential predictors x1 and x2 with α = 0.05 for tests, which model would you expect to be selected? Why?