2806 04042 or 16848 08764 from the table above

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: iginal Data 60 Linear Regression w/o outlier 50 40 30 20 10 0 0 c) 10 20 30 40 Deformation (mm) The original data and linear regression are plotted. Our assumption about the outlier is valid. Give 90% confidence intervals for the two parameters you found in (b). From the table above, we can get i 9/19 CEE 202 – Engineering Risk & Uncertainty Spring 2011 (Bond) n 2 H Homework 11 Diff. between Means; Least ­Squares 2 SST = ( n − 1) S = ∑ ( H i − H ) = 1046 i =1 n ˆ SSE = ∑ H i − H i i =1 ( n − 1) S 2 D ( = 53.56 2 = ∑ ( Di − D ) = 605.20 i =1 ∑(D − D)(H i − H ) = −775 i i =1 n ∑(H i =1 s= 2 n n ) n 2 i − H ) − b ∑ ( Di − D ) ( H i − H ) i =1 n−2 1046 − ( −1.2806 ) × ( −775 ) = 4.2253 5−2 = Or you can use SSE 53.56 s= = = 4.2253 n−2 3 α = 1 − 0.9 = 0.1 , in each tail α 2 = 0.05 ; d.f.=n ­2=5 ­2=3 Using Table A4. or function tinv in Excel, we find that t0.05 = 2.353 for 3 degrees of freedom. Therefore, a 90% confidence interval for b is s 4.2253 β = b ± tα /2 = −1.2806 ± 2.353 × = n 605.20 2 ∑ ( Di − D ) i =1 −1.2806 ± 0.4042 or ( −1.6848, −0.8764 ) From the table above, we can also get n 2 i ∑D = 2335 , i =1 A 90% confidence interval for a is n s α = a ± tα /2 2 i ∑D i =1 n n∑ ( Di − D ) = 74.8186 ± 2.353 × 2 4.2253 2335 = 5 × 605.2 i =1 74.8186 ± 8.7349or(66.0837,83.5534) 10/19 CEE 202 – Engineering Risk & Uncertainty Spring 2011 (Bond) d) e) Homework 11 Diff. between Means; Least ­Squares What is the coefficient of determination for the relationship you determined in (b)? SSE 53.56 The coefficient of determination: R 2 = 1 − = 1− = 0.9488 SST 1046 What is the correlation coefficient between H and D? The correlation coefficient between H and...
View Full Document

Ask a homework question - tutors are online