 1 
Solutions to BUAD 310 InClass Practice
11/3/09
Regression Analysis: pagecost versus circ
The regression equation is
pagecost =
21612 + 4.9268 circ
Predictor
Coef
SE Coef
T
P
Constant
21612
2568
8.42
0.000
circ
4.9268
0.5342
9.22
0.000
S =
13757.0
RSq =
64.9%
Analysis of Variance
Source
DF
SS
MS
F
P
Regression
1
16098031920
16098031920
85.06
0.000
Residual Error
46
8705772362
189255921
Total
47
24803804282
Predicted Values for New Observations
New
Obs
Fit
SE Fit
95% CI
95% PI
1
46246
2243
(41731, 50760)
(18189, 74303)
Values of Predictors for New Observations
New
Obs
circ
1
5000
Questions:
1.
What is the leastsquares regression equation to predict Pagecost from Circulation?
pagecost =
21612 + 4.9268 circ
2.
Is Circulation a statistically useful variable in predicting Pagecost? Justify your answer.
Test H0: β1=0 vs H
a
: β1≠0
pvalue =0.000. Reject at 5% level.
So at 5% level, Circulation is a statistically useful variable
in predicting Pagecost
3.
Is it likely that the regression line goes through the origin?
Test H0: β0=0 vs H
a
: β0≠0
pvalue =0.000. Reject at 5% level.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document 2 
So it is not likely that the regression line goes through the
origin.
4.
What percent of variation in Pagecost can be explained by Circulation?
RSq = 64.9%
5.
Is it possible that the intercept could be 20,000?
Test H0: β0=
20,000 vs Ha
: β0≠
20,000
t=(2161220000)/ 2568=0.6277, df=46
pvalue = 2P(T>0.6277)>0.1
At 10% level, it is possible that the intercept is 20,000
6.
What is the estimate of the standard deviation of the error term in the simple linear regression model?
S = 13757.0
This is the end of the preview.
Sign up
to
access the rest of the document.
 '07
 Lv
 Business, Regression Analysis, Errors and residuals in statistics, Source Regression Residual, Variance Source Regression, pagecost

Click to edit the document details