ISE 426-526 - Lecture 14 - MLR Part II - B&amp;W

# ISE 426-526 - Lecture 14 - MLR Part II - B&amp;W - ISE...

This preview shows pages 1–7. Sign up to view the full content.

1 ISE 426-526 Lecture 14 – 2007 Regression Analysis – Part II Model Diagnostics

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 PRESS Statistic R 2 for Prediction
3 Lack of Fit • Questions the linear model • Requires repeat observations • Repeat observations are used to obtain an estimate of s 2 • Partitions the residual SS Lack of Fit – cont. 2 ˆ () E ij i SS yy =− y ij = j th observation of the response at x i i = 1,2,…m j = 1, 2, …n 2 11 mn PE ij i ij SS y y == ∑∑ 2 1 ˆ m LOF i i i i SS n y y = If fitted ŷ i is close to the corresponding average ÿ i then there is strong evidence that the regression function is linear. If not it is likely that the regression function is not linear.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 Lack of Fit – cont. ANOVA for Lack of Fit
5 Decision Rule LOF - Example x i y i df 1.3 2.3, 1.8 2.05 2.0 2.8, 1.5 2.15 2.7 2.2 3.3 3.8, 1.8 2.80 3.7 3.7, 1.7 2.70 4.0 2.8, 2.8, 2.2 2.60 4.7 5.4, 3.2, 1.9 3.50 5.0 1.8 5.3 3.5, 2.8, 2.1 2.80 5.7 3.4 6.0 3.2, 3.0 3.10 6.3 3 6.7 5.9

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
6 LOF – Example – cont. LOF – Example – ANOVA. The regression equation is yi = 1.44 + 0.338 xi Predictor Coef SE Coef T P Constant 1.4364 0.5900 2.43 0.023 xi 0.3379 0.1319 2.56 0.018 S = 0.981503 R-Sq = 23.0% R-Sq(adj) = 19.5% PRESS = 25.2416
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 8

ISE 426-526 - Lecture 14 - MLR Part II - B&amp;W - ISE...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online