The following table presents the weight (X1), age (X2), and plasma lipid levels of total cholesterol (Y) for a hypothetical sample of 25 patients suffering from hyperlipoproteinemia, before drug therapy.
a. Three estimated regression models, along with their sums-of-squares results, are as follows:

For each of these models, determine the predicted cholesterol level (Y) for patient 4, and compare these predicted cholesterol levels with the observed values. Comment on your findings.

b. Determine R2-values for each of the three models considered in part (a). If you use R2 and model simplicity as selection criteria, which model appears to be the best predictive model?

Transcribed Image Text:

Y= 77.9831 0.41TXİ + 5.217X2 SSY= 145,377.0400, SSE= 42,806.2254 Y= 199.2975 + 1.622% Y 102.5751 5.321XSSE 43,444.3743 SSE-135,145.3138 23 60 48 34 51 50 34 46 23 37 40 30 57 59 69 60 2 5 785 or g-23 15 16 17 18 19 20 21 22 23 24 25 5 235 73 65 70 76 69 63 72 79 75 27 7677667 300 198 995 400 360 525 190 105 360 1234567890123