Scribd
Upload
57 views5 pages

Weekly Usage Hrs Annual Maintenance Expense (1000s)

The document presents the results of a linear regression analysis examining the relationship between weekly usage hours and annual maintenance expense in thousands of dollars for 10 observations. The regression results in a strong positive correlation (R^2 = 0.856) where 85% of the variation in annual maintenance expense is explained by weekly usage hours. The regression equation derived is: Annual Maintenance Expense = 10.528 + 0.953 * Weekly Usage Hours. With a low p-value of 0.0001, the analysis finds a statistically significant relationship between weekly usage hours and annual maintenance expense.

Uploaded by

Annapurna Purohit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
57 views5 pages

Weekly Usage Hrs Annual Maintenance Expense (1000s)

The document presents the results of a linear regression analysis examining the relationship between weekly usage hours and annual maintenance expense in thousands of dollars for 10 observations. The regression results in a strong positive correlation (R^2 = 0.856) where 85% of the variation in annual maintenance expense is explained by weekly usage hours. The regression equation derived is: Annual Maintenance Expense = 10.528 + 0.953 * Weekly Usage Hours. With a low p-value of 0.0001, the analysis finds a statistically significant relationship between weekly usage hours and annual maintenance expense.

Uploaded by

Annapurna Purohit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Weekly usage Hrs Annual maintenance expense ( 1000s)

13 17
10 22
20 30
28 37
32 47
17 30.5
24 32.5
31 39
40 51.5
38 40

Mean (Y bar) 34.65

Sum of Squared Error (SSE) 144.47406252

Weekly Usage H
( 1000s)
The sqaure of sum of the difference between the observed value of the dependent variable and the value predicted using
equation is Sum of Sqaures due to Error. Here we have calculated predicted value (Y^) by putting
60 the values of weekly us
regression equation. Residuals have been found out by subtracting predicted value from observed value, i.e., (Yi-Y^). Finally

ANmual maintenance expense (1000s)


of all the residuals will give us SSE i.e., summation of sqaure of (Yi-Y^).
50
f(x) = 0.953440
Total Sum of Sqaures (SST) 1004.525 40

30

The sqaure of the sum of the difference between the observed value of the dependent varaible 20 and the mean value of th
variable is Total Sum of Sqaures (SST). Difference has been found out by subtracting mean value from the observed value,
Then, sum of sqaure of this difference will give us SST i.e., summation of sqaure of (Yi-Y bar).
10

Sum of Sqaures Due to Regression (SSR) 859.97800252 0


5 10 15

The sqaure of the sum of the difference between the predicted value of the dependent varaible and the mean value of th
variable is Total Sum of Sqaures (SST). Difference has been found out by subtracting mean value from the predicted value,
Then, sum of sqaure of this difference will give us SSR i.e., summation of sqaure of (Y^-Y bar).

Also, SST= SSR+SSE

R^2= SSR/SST 0.856104131325751


R^2 is the proportion of the variation in y values that is explained by the linear relationship with x. So here 85 percent of va
maintenance expense is explained by weekly usage hours. Rest 15 percent of variation in annual maintenance expense is
some other variable.
Y^ Residuals Error (Squared Residuals) (Yi-Y bar)
22.9222 -5.9222 35.07245284 -17.65
20.062 1.938 3.755844 -12.65
29.596 0.404 0.163215999999997 -4.65
37.2232 -0.2232 0.049818239999999 2.35
41.0368 5.9632 35.55975424 12.35
26.7358 3.7642 14.16920164 -4.15
33.4096 -0.9096 0.827372159999995 -2.15
40.0834 -1.0834 1.17375556000001 4.35
48.664 2.836 8.04289599999999 16.85
46.7572 -6.7572 45.65975184 5.35

SSE 144.47406252 SST

Weekly Usage Hrs vs Annual maintenance expense


( 1000s)
ent variable and the value predicted using the regression
alue (Y^) by putting
60 the values of weekly usage hrs in the
ue from observed value, i.e., (Yi-Y^). Finally, sum of sqaure
ANmual maintenance expense (1000s)

tion of sqaure of (Yi-Y^).


50
f(x) = 0.95344043969982 x + 10.5279568755946
40

30

pendent varaible
20 and the mean value of the dependent
cting mean value from the observed value, i.e., (Yi-Y bar).
summation of sqaure of (Yi-Y bar).
10

0
5 10 15 20 25 30 35 40 45
Weekly usage hrs

ependent varaible and the mean value of the dependent


ting mean value from the predicted value, i.e., (Y^-Y bar).
summation of sqaure of (Y^-Y bar).

lationship with x. So here 85 percent of variation in annual


ariation in annual maintenance expense is dependent on
(Yi-Y bar)^2 (Y^-Y bar) (Y^-Y bar)^2
311.5225 -11.7278 137.54129284
160.0225 -14.588 212.809744
21.6225 -5.054 25.542916
5.5225 2.5732 6.62135824
152.5225 6.3868 40.79121424
17.2225 -7.9142 62.63456164
4.6225 -1.2404 1.53859216
18.9225 5.4334 29.52183556
283.9225 14.014 196.392196
28.6225 12.1072 146.58429184

1004.525 SSR 859.97800252

expense

40 45
SUMMARY OUTPUT

Regression Statistics
Multiple R 0.925298194552821
R Square 0.856176748842709
Adjusted R Square 0.838198842448048
Standard Error 4.24961838534911
Observations 10

ANOVA
df SS MS F
Regression 1 860.050948631223 860.0509 47.623829496
Residual 8 144.474051368777 18.05926
Total 9 1004.525

Coefficients Standard Error t Stat P-value


Intercept 10.5279568755945 3.74486728221436 2.811303 0.0227972449
Weekly usage Hrs 0.95344043969982 0.138159709382233 6.901002 0.0001244026

Calculation of MS (Using formulae) Mean Square Error(MSE) is sqaure roo


MS = SS/ degree of freedom
MSE
MS (Regression) 860.050948631223
MS (Residual) 18.0592564210972 F= MSR/MSE

Now using F- distribution table and the

Null Hypothesis: There is no significant relation between weekly usage and annual maintenance hours
Alteernate Hypothesis: There is significant relation between weekly usage and annual maintenance hours

p-value 0.00012440260293

Since p-value is less than 0.05, we reject null hypothesis. It means that there is stastically significant relation between week
Significance F
0.0001244026

Lower 95% Upper 95% Lower 95.0% Upper 95.0%


1.892277437 19.16363631417 1.892277437 19.1636363142
0.6348435785 1.272037300854 0.634843579 1.27203730085

an Square Error(MSE) is sqaure root of MS Residual

4.2496183853

47.623829496

w using F- distribution table and the degree and freedom we get the p-value.

maintenance hours
al maintenance hours

significant relation between weekly usage and annual maintenance hours

You might also like