In: Statistics and Probability
The number of cell phones per 100 residents in countries in Europe is given in table #9.3.9 for the year 2010. The number of cell phones per 100 residents in countries of the Americas is given in table #9.3.10 also for the year 2010 ("Population reference bureau," 2013). Find the 98% confidence interval for the different in mean number of cell phones per 100 residents in Europe and the Americas.
Table #9.3.9: Number of Cell Phones per 100 Residents in Europe
100 |
76 |
100 |
130 |
75 |
84 |
112 |
84 |
138 |
133 |
118 |
134 |
126 |
188 |
129 |
93 |
64 |
128 |
124 |
122 |
109 |
121 |
127 |
152 |
96 |
63 |
99 |
95 |
151 |
147 |
123 |
95 |
67 |
67 |
118 |
125 |
110 |
115 |
140 |
115 |
141 |
77 |
98 |
102 |
102 |
112 |
118 |
118 |
54 |
23 |
121 |
126 |
47 |
Table #9.3.10: Number of Cell Phones per 100 Residents in the Americas
158 |
117 |
106 |
159 |
53 |
50 |
78 |
66 |
88 |
92 |
42 |
3 |
150 |
72 |
86 |
113 |
50 |
58 |
70 |
109 |
37 |
32 |
85 |
101 |
75 |
69 |
55 |
115 |
95 |
73 |
86 |
157 |
100 |
119 |
81 |
113 |
87 |
105 |
96 |
b.) If needed, state the null and alternative hypotheses and the level of significance and don't "bump" them into one line (refer to the textbook)
Ho :
HA :
a =
c.) State and check the assumptions for a hypothesis test
d.) Find the sample statistic, test statistic, and p-value, confidence interval
e.) Conclusion
Since the p-value……..fail to reject or reject H o or confidence intervals...
f.) Interpretation ( statistical and real world)
Solution
Final answers in the stipulated format are given below. Back-up Theory and Details of calculations follow at the end.
Part (a)
Random variables
X = Number of Cell Phones per 100 Residents in Europe
Y = Number of Cell Phones per 100 Residents in Americas Answer 1
Parameter in words
λ1 = average number of Cell Phones per 100 Residents in Europe
λ2 = average number of Cell Phones per 100 Residents in Americas Answer 2
Part (b)
Hypotheses:
Null H0 : λ1 = λ2 Vs HA : λ1 ≠ λ2 Answer 3
Part (c)
Assumption
Difference between estimates of λ1 and λ2 follows N(0, 1). Answer 4
Validity:
Since the sample sizes are large enough, by Central Limit Theorem, the above assumption holds valid. Answer 5
Part (d)
Sample statistic: estimates of λ1 and λ2 108.15 and 87.21 respectively. Answer 6
Test statistic: Z = = (m1 – m2)/√[m{(1/n1) + (1/n2)}] = 9.9646 Answer 7
where m1 and m2 are averages based on n1 and n2 observations respectively and
m = (n1m1 + n2m2)/( n1 + n2)
p-value < 0.0001 Answer 8
98% Confidence interval: [16.40, 26.19] Answer 9
Part (e)
Conclusion
Since the p-value is less than the significance level we reject H0 or confidence interval does not hold zero Answer 10
Part (f)
Interpretation (statistical): λ1 ≠ λ2 Answer 11
Interpretation (real world)
The mean number of Cell Phones per 100 Residents in Europe is different from that in Americas. Answer 12
DONE
Back-up Theory and Details of calculations
Test
Let X = Number of Cell Phones per 100 Residents in Europe
Y = Number of Cell Phones per 100 Residents in Americas
Then, X ~ Poisson (λ1) and Y ~ Poisson (λ2)
Hypotheses:
Null H0 : λ1 = λ2 Vs HA : λ1 ≠ λ2
Test Statistic:
Let m1 = sample estimate of λ1 and m2 = sample estimate of λ2
Z = (m1 – m2)/(√m) if m1 and m2 are based on single observations and m = (m1 + m2)/2
= (m1 – m2)/√[m{(1/n1) + (1/n2)}] if m1 and m2 are averages based on n1 and n2 observations
respectively and m = (n1m1 + n2m2)/( n1 + n2)
Calculations:
n1 |
53 |
n2 |
39 |
m1 |
108.1509 |
m2 |
87.2051 |
Zcal |
9.9646 |
α |
0.01 |
Zcrit |
2.5758 |
p-value |
0.00E+00 |
1 - (α/2) |
0.995 |
1- α |
0.99 |
1-pvalue |
1.00E+00 |
(1/n1) |
0.0189 |
(1/n2) |
0.0256 |
sum |
0.0445 |
sqrt(msum) |
2.1020 |
n1m1+n2m2 |
9133 |
n1+n2 |
92 |
m |
99.2717 |
Distribution, Critical Value, p-value:
Under H0 , distribution of Z can be approximated by N(0, 1). Hence, given α as the level of significance,
Critical Value = upper (α/2)% point of N(0, 1) and p-value = P(Z > | Zcal |).
Using Excel Functions: Statistical NORMSINV and NORMSDIST the above are found to be as shown in the above table.
Decision Criterion (Rejection Region):
Reject H0, if | Zcal | > Zcrit or p-value < α.
Decision:
Since | Zcal | > Zcrit or equivalently p-value < α, H0 rejected.
Conclusion:
There is enough evidence to suggest that the mean number of Cell Phones per 100 Residents are different between Europe and Americas..
Confidence interval
100(1 - α) % Confidence Interval for (λ1 - λ2) is: (Mean X – Mean Y) ± (Zα/2)√[m{(1/n) + (1/m)}]
where m = (nMeanX + mMeanY)/(n + m), where
Zα/2 is the upper (α /2)% point of N(0, 1), n and m are the two sample sizes.
Calculations
n |
53 |
m |
39 |
m1 |
108.5 |
m2 |
87.21 |
m |
99.47489 |
d = m1 - m2 |
21.29 |
α |
0.02 |
1 - (α/2) |
0.99 |
Zα/2 |
2.3263 |
(1/n)+(1/m) |
0.0445 |
m{(1/n) + (1/m)} |
4.4275 |
sqrt |
2.1042 |
MoE |
4.8950 |
LB |
16.3950 |
UB |
26.1850 |