This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: HW Set 4: Due 4/27 in quiz session. HW9: 3.21(c,d; use the print-out given in the problem as much as possible.), 3.22 (By computer. The statement "assess the quality of the fit, ..." means "Compute R-squared and interpret it.") 3.25(b,c,d; by hand) . HW10: hw_P, hw_Q, hw_R (end of lecture 12; all by computer). HW11: hw_S and hw_T (lecture 13). Graded problems (and points): TBA. 3.21 (c) The least squares line is y ˆ = 3.2925 + .10748x. For a beam whose modulus of elasticity is x = 40, the predicted strength would be y ˆ = 3.2925 + .10748(40) = 7.59. The value x = 100 is far beyond the range of the x values in the data, so it would be dangerous (i.e., potentially misleading) to extrapolate the linear relationship that far. (d) From the printout, SSResid = 18.736, SSTo = 71.605, and the coefficient of determination is r 2 = .738 (or, 73.8%) . The r 2 value is large, which suggests that the linear relationship is a useful approximation to the true relationship between these two variables. 3.22 Using Minitab to run this regression, we obtain the following least squares regression line: x y 00882 . 86 . 2 ˆ + = 40 50 60 70 80 90 100 110 3.2 3.3 3.4 3.5 3.6 3.7 3.8 thickness wavelength Regression Plot 96.1% of the observed variation in PL wavelength can be attributed to the 96....
View Full Document
This document was uploaded on 05/15/2010.
- Spring '08