(b) Regardless of your proposed model in (a), consider model where 1, if categor

Question

(b) Regardless of your proposed model in (a), consider model where 1, if category 1 1, if category 2 The Minitab outputs for this model are shown below Predictor Constant Coef SE Coef -0.8365 0.5518 -1.52 0.133 0.96327 0.06980 13.80 0.000 0.1502 0.7522 0.20 0.842 1.7934 0.77322.32 0.023 -0.10888 0.01341 -8.12 0.000 0.06482 0.09948 0.65 0.516 0.06783 0.08818 0.77 0.444 0.00630 0.01829 0.34 0.731 0.09402 0.01780 5.28 0.000 z1*x2 22 x 2 S. 1.88621 R-Sq 95.3% R-Sq ( adj ) 94.9% - . Analvsis of Variance DF Source Regression Residual Error 91 323.76 Total MS 8 6586.58 823.32 3.56 99 6910.34

Explanation / Answer

iii.

Yes, we can use backward elimination here. We ill remove the predictor with highest p-value greater than critical value of signifance level (0.05). The highest p-value is 0.842 for variable z1. So, variable  z1 should be removed first.

Related Questions

Navigate

• Browse (All)
• Browse #
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.