Question

The following data were collected on a simple random sample of 20 patients with hypertension: Y=mean arterial blood pressure (mmHg), X1=age(years),  X2= weight (kg), X3=body surface area (sq m), X4=duration of hypertension, X5 =basal pulse (beats/min), X6=measure of stress. A researcher is interested in developing a regression model to predict mean arterial blood pressure and has produced the following output:

> rcorr(as.matrix(hyper))

      Y   X1  X2   X3   X4  X5   X6

Y  1.00 0.66 0.95 0.87 0.29 0.72 0.16

X1 0.66 1.00 0.41 0.38 0.34 0.62 0.37

X2 0.95 0.41 1.00 0.88 0.20 0.66 0.03

X3 0.87 0.38 0.88 1.00 0.13 0.46 0.02

X4 0.29 0.34 0.20 0.13 1.00 0.40 0.31

X5 0.72 0.62 0.66 0.46 0.40 1.00 0.51

X6 0.16 0.37 0.03 0.02 0.31 0.51 1.00

n= 20

a. Which X variable has the weakest relationship with Y?

b. Which X variable has the strongest relationship with Y?

c. As we discussed, it is preferable to include predictor (X) variables that are not highly correlated. Which 2 predictor variables seem to be highly correlated?

Answer 1

ANSWER :

A). From the matrix, it is clear that X6 has the weakest relationship with Y.

B). X2 has the strongest relationship.

C). X2 and X3 seems to be highly correlated (0.88) (from the matrix).