This preview shows page 1. Sign up to view the full content.
Unformatted text preview: SSE = (11) – (10)
ˆ
= ∑ (y  y )2
= (n1)sy2
= SSM/1 = (9)/(6)
= SSE/(n2) = (10)/(7) = se2
= MSM/MSE = (12)/(14)
= Pr (F>Fstat), where Fstat = (14)
= ybar – b1 (xbar), where b1 = (17)
= r(sy/sx)
2 1
x
+
n ∑ xi − x 2 (18) = SE(b0) = se (19) = SE(b1) = (20)
(21)
(22)
(23)
(24)
(25)
(26)
(27) = (16)/(18)
= (17)/(19)
= 2 x Pr (tn2 > (20))
= 2 x Pr (tn2 >(21))
= (16) – t*n2 (18)
= (17)  t*n2 (19)
= (16) + t*n2 (18)
= (17) + t*n2 (19) ( ) se
2 ∑ (x − x )
i *** END OF ASSIGNMENT 6 *** (or textbook version) (or textbook version) Significance F (15) Lower 95% (24) (25) Upper 95% (26) (27) Marking Scheme (Total = 20):
Question 1. (10 marks)
a) Scatterplot and regression line – 1 mark
Assumptions check – 2 marks (0.5 for each)
b) Regression equation – 1 mark
c) Hypothesis test – 2 marks: 0.5 for hypotheses, 1 for test statistic, 0.5 for conclusion
d) CI – 2 marks: 1 for estimated mean, 1 for margin of error
e) PI – 2 marks: 1 for estimated mean, 1 for margin of error
Question 2. (8 marks)
a) Fitted regression model – 2 marks
b) R square – 1 mark (0.5 for value, 0.5 for interpretation)
c) Interpretation – 1 mark (accept any reasonable answer)
d) Final model – 2 marks
e) Check on assumptions – 2 marks (two residual plots needed and either the normal probability plot or a
histogram; 1 mark for plots, 1 mark for conclusions)
Question 3. (2 marks)
If the template is filled out, give full marks. Just do a spot check on a few entries of your choice....
View
Full
Document
This note was uploaded on 01/16/2014 for the course COMM 291 taught by Professor E.fowler during the Fall '10 term at The University of British Columbia.
 Fall '10
 E.Fowler

Click to edit the document details