In: Statistics and Probability
Multiple regression is used by accountants in cost analysis to shed light on the factors that cause costs to be incurred and the magnitudes of their effects.
A petroleum company wanted to evaluate different blends of its gasoline - made by inserting different additives into the manufacturing process. To test this the firm acquired thirty identical cars (make and model) from an auto company and randomly assigned the cars to one of the three blends. The cars were then driven for 30 eight hour days on a dynamometer simulating city and highway driving conditions and the mileage obtained (MPG) was measured at the conclusion of the study. The results are below and require analysis to obtain a decision as to whether one of the blends or more was superior to the others.
Gasoline Blends
Blend A | Blend B | Blend C |
28.6 | 25.6 | 29.6 |
27.6 | 26.6 | 28.6 |
31 | 31 | 32.5 |
29.2 | 29.4 | 29.8 |
30.2 | 29.5 | 29.6 |
28.9 | 30 | 29.4 |
29.1 | 29.6 | 32 |
30.2 | 27.5 | 31.2 |
28.6 | 31.1 | 31.9 |
28.4 | 29.1 | 28.2 |
A | B | C | ||||
count, ni = | 10 | 10 | 10 | |||
mean , x̅ i = | 29.180 | 28.94 | 30.28 | |||
std. dev., si = | 1.014 | 1.817 | 1.507 | |||
sample variances, si^2 = | 1.028 | 3.303 | 2.271 | |||
total sum | 291.8 | 289.4 | 302.8 | 884 | (grand sum) | |
grand mean , x̅̅ = | Σni*x̅i/Σni = | 29.47 | ||||
square of deviation of sample mean from grand mean,( x̅ - x̅̅)² | 0.082 | 0.277 | 0.662 | |||
TOTAL | ||||||
SS(between)= SSB = Σn( x̅ - x̅̅)² = | 0.822 | 2.774 | 6.615 | 10.21067 | ||
SS(within ) = SSW = Σ(n-1)s² = | 9.256 | 29.724 | 20.436 | 59.4160 |
no. of treatment , k = 3
df between = k-1 = 2
N = Σn = 30
df within = N-k = 27
mean square between groups , MSB = SSB/k-1 =
5.1053
mean square within groups , MSW = SSW/N-k =
2.2006
F-stat = MSB/MSW = 2.3200
anova table | 0.146648794 | |||||
SS | df | MS | F | p-value | F-critical | |
Between: | 10.21 | 2 | 5.11 | 2.32 | 0.1176 | 3.35 |
Within: | 59.42 | 27 | 2.20 | |||
Total: | 69.63 | 29 |
F =2.32
p value = 0.1176
α =
0.05
conclusion : p-value>α , do not reject null
hypothesis
So, none of them superior to the others.
..............
THANKS
revert back for doubt
please upvote