In: Statistics and Probability
1. a) Suppose average monthly sales for retail locations across the United States are approximately normally distributed with variance σ^2= 5200. We took a sample of size 50 and found ̄x= 12018, Using this, conduct a hypothesis test with α= 0.05 to test the null hypothesis that the mean is 12000 vs. the alternative hypothesis that it is not. For full credit, state the null and alternative hypothesis, the test statistic, the rejection region, and your conclusion.
b)Using the setup from part a, if we know that the true mean is 12030, what is the probability of a type II error?
c)Using the setup from part a, what would be the p-value of this test? Would you reject the null hypothesis if α= 0.01? How about if α= 0.1
d)Using the set up from part a, perform the hypothesis test again, but now use the alternative hypothesis that the mean is actually greater than 12000.
e)Using the setup from part a, if we know the true mean is 12030 and we want the probability of a type I error to be 0.05 and the probability of a type II error
to be 0.10, what is the minimum sample size required to ensure this?
a)
Ho : µ = 12000
Ha : µ ╪ 12000
(Two tail test)
Level of Significance , α =
0.05
population std dev , σ =
72.1110
Sample Size , n = 50
Sample Mean, x̅ = 12018.0000
Standard Error , SE = σ/√n = 72.1110 / √
50 = 10.1980
Z-test statistic= (x̅ - µ )/SE = (
12018.000 - 12000 ) /
10.1980 = 1.77
critical z value, z* = ± 1.9600
[Excel formula =NORMSINV(α/no. of tails) ]
rejection region: |test stat | > 1.96
decision: |test stat| < critical value , so do not reject Ho
conclusion: mean is 12000
...........
b)
true mean , µ =
12030
hypothesis mean, µo = 12018
significance level, α = 0.05
sample size, n = 50
std dev, σ = 72.111
δ= µ - µo = 12
std error of mean, σx = σ/√n =
72.1110 / √ 50 =
10.19804
Zα/2 = ± 1.960 (two tailed
test)
We will fail to reject the null (commit a Type II error) if we get
a Z statistic between
-1.960
and 1.960
these Z-critical value corresponds to some X critical values ( X
critical), such that
-1.960 ≤(x̄ - µo)/σx≤ 1.960
11998.012 ≤ x̄ ≤ 12037.988
now, type II error is ,ß = P
( 11998.012 ≤ x̄ ≤
12037.988 )
Z = (x̄-true
mean)/σx
Z1 = (
11998.012 - 12030 )
/ 10.19804 = -3.137
Z2 = (
12037.988 - 12030 )
/ 10.19804 = 0.783
so, P( -3.137 ≤ Z
≤ 0.783 ) = P ( Z ≤
0.783 ) - P ( Z ≤ -3.137
)
= 0.783
- 0.001 = 0.78241 [
Excel function: =NORMSDIST(z) ]
ß = 0.78241
................
c)
p-Value = 0.0776
α= 0.01
p value > 0.01, do not reject Ho
if α= 0.1
p value < 0.1, reject Ho
...
d)
Ho : µ = 12000
Ha : µ > 12000
(Right tail test)
Level of Significance , α =
0.05
population std dev , σ =
72.1110
Sample Size , n = 50
Sample Mean, x̅ = 12018.0000
Standard Error , SE = σ/√n = 72.1110 / √
50 = 10.1980
Z-test statistic= (x̅ - µ )/SE = (
12018.000 - 12000 ) /
10.1980 = 1.77
critical z value, z* =
1.6449 [Excel formula =NORMSINV(α/no. of tails)
]
Decision: test stat > critical value , Reject null
hypothesis
..............
e)
True mean µ = 12030
hypothesis mean, µo = 12018
Level of Significance , α =
0.05
std dev = σ = 72.111
power = 1-ß = 0.9
ß= 0.1
δ= µ - µo = 12
Z ( α ) = 1.6449 [excel
function: =normsinv(α)
Z (ß) = 1.2816 [excel
function: =normsinv(ß)
sample size needed = n = ( ( Z(ß)+Z(α) )*σ / δ )² = (
( 1.2816 + 1.6449 )
* 72.1 / 12 ) ²
= 309.2498
so, sample size =
310.000
thanks
revert back for doubt