In: Statistics and Probability
Q1: A transport company wants to compare the fuel efficiencies of the two types of lorry it operates. It obtains data from samples of the two types of lorry, with the following results:
Type |
Average mpg |
Std devn |
Sample size |
A |
30.1 |
7.6 |
13 |
B |
33.2 |
5.8 |
17 |
At 90% confidence level, test the hypothesis that lorries of type B are more efficient than type A. Use critical value approach and assume unequal variation in mpg of two types.
Solution
Let X = mpg of lorry of Type A and Y = mpg of lorry of Type B
Then, X ~ N(µ1, σ12) and Y ~ N(µ2, σ22), where σ12 and σ22 are unequal [given, ‘assume unequal variation in mpg of two types’ ] and unknown.
Claim:
Lorries of Type B are more efficient than lorries of Type A
Hypotheses:
Null: H0: µ1 = µ2 Vs Alternative: HA: µ1 < µ2
Test Statistic:
t = (Xbar - Ybar)/√[{(s12/n1) + (s22/n2)}] where
Xbar and Ybar are sample averages and s1,s2 are sample standard deviations based on n1 observations on X and n2 observations on Y respectively.
Calculations
Summary of Excel calculations is given below:
n1 = |
13 |
n2 = |
17 |
Xbar = |
30.1 |
Ybar = |
33.2 |
s1 = |
7.6 |
s2 = |
5.8 |
s1^2/n1 |
4.443077 |
s2^2/n2 |
1.978824 |
Sum S |
6.4219 |
sqrt(S) |
2.534147 |
Xbar - Ybar |
-3.1 |
tcal |
-1.22329 |
α |
0.1 |
DF - ν |
22 |
tcrit |
-1.32319 |
ν-calculation |
|
S1 = s1^2/n1 |
4.443077 |
S2 = s2^2/n2 |
1.978824 |
S = S1 + S2 |
6.4219 |
ν1 |
12 |
ν2 |
16 |
F1 |
2028 |
F2 |
4624 |
D1= s1^4/F1 |
1.645078 |
D2= s2^4/F2 |
0.244734 |
D = D1 + D2 |
1.889812 |
ν = S^2/D |
21.82271 |
[ν] |
22 |
Distribution, Significance Level , Critical Value
Under H0, t ~ tν, where
ν = {(s12/n1) + (s22/n2)}/{(s14/n12x ν1) + (s24/n22x ν2)}; ν1 = n1 - 1 and ν2 = n2 - 1
Hence, for level of significance α%, Critical Value = lower α% point of tν
Using Excel Functions, Statistical TINV, the above are found to be as shown in the above table: [given 90% confidence level, α = 10%, i.e., 0.1]
Decision:
Since tcal > tcrit, H0 is accepted.
Conclusion:
There is NOT sufficient evidence to suggest that the claim is valid. i.e.,
Lorries of Type B are NOT more efficient than lorries of Type A. ANSWER
DONE