Girolando cattle weight. A group of researchers generated a model of the body measurements of Girolando cattle by extracting their image measurements, aiming to understand which measurements further explain the body weight of the cattle. The research was published in Brazilian Journal of Animal Science (Vol. 49, 2020). The potentially important predictors of body weight measurement 1y2 (kg) are listed in the accompanying table. x1 = hip width (cm) x2 = body length (cm) x3 = tail distance to the neck x4 = dorsum area (cm2 ) x5 = dorsum perimeter (cm) x6 = wither height (cm) x7 = hip height (cm) x8 = area of the body lateral area (cm2 ) x9 = perimeter of the lateral area (cm2 ) x10 = rib height (cm)

a. The researchers discovered that the variable hip width (x1) was highly correlated with the other potential independent variables. Therefore, hip width was dropped from the model. Do you agree with this decision? Explain.

b. The researchers used stepwise regression on the remaining nine potential independent variables in order to search for a parsimonious set of predictor variables. Do you agree with this decision? Explain.

c. The stepwise regression selected the following independent variables: x1 = hip width and x4 = dorsum area. These variables were statistically significant in the stepwise model, with R2 = .70. Consequently, the researchers used the estimated stepwise model to predict body weight. Do you agree with this decision? Explain.