0% found this document useful (0 votes)
8 views6 pages

Statistical Hypothesis Testing Guide

Uploaded by

aashima chopra
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
8 views6 pages

Statistical Hypothesis Testing Guide

Uploaded by

aashima chopra
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Solution

Solution 1:

Assumption

a. Samples are randomly selected from the population.

Step 1 : Hypotheses:
H 0 : μ=3 (null hypothesis: the mean number of cavities is 3, suggesting the
toothpaste doesn't have an effect

(Alternative hypothesis H 1 : μ <3: the mean number of cavities is less than 3,


suggesting the toothpaste does have an effect)

Given:

 n = 2500 (sample size)


 ˉxˉ = 2.95 (sample mean)

 μ = 3 (population mean)
 σ = 1 (population standard deviation)
 Step 2: α=0.05

Step 3 : Using the formula for the z-test statistic: zobs= 50 *(2.95-3)/1=-
2.5

Step 4: Z_critical=NORM .[Link](0.05)(-[Link](0.95)(Using


Symmentery))=-1.645

Calculate the p -value, since it is a left tailed test p-value


=P(Z<zobs)=0.00621

([Link](-2.5,0,1,1)

 Step5 :zobs< Z_critical ( p-value <0.05)

 Step 6 : We should reject the null hypothesis at 5% level of significance.

Conclusion : At the 5 percent level of significance, the data is strong


enough to establish the claim of the toothpaste advertisement. The study

suggests that the new toothpaste reduces the mean number of cavities
for children in their cavity-prone years.

Alternative : Critical value Approach :

For a left-tailed test with α=0.05, the critical z-value is -1.645


([Link](0.05,0,1)

zobs< the critical z-value. We should reject the null hypothesis at 5%


level of significance.

2. H 0 : μ ≥10 vs H 1 : μ <10

(Note : In the null hypothesis, , we perform our test procedures at the


boundary points whenever there is ≥∨≤ sign. We will calculate our test
statistic at the boundary point i.e at μ=10 . Because if we reject our null
hypothesis at μ=10 , with the alternative : μ<10 , it will automatically leads
to the rejection of the null hypothesis for μ>10 . Similar things will
happen when we fail to reject. So even if we incorporate ≥∨≤ , to
complete the hypothesis, we will proceed similarly like “=” in the null
hypothesis.)

3. Assumption :

a) Population distribution is normal ( The sample size is small)

b) Random samples are drawn from the population.

. H 0 : μ ≥200 vs H 1 : μ <200
Test Statistic : zobs=sqrt(8)* (199.125-200)/5=-0.4949

p-value =P(z<-0.4949)=0.31 ([Link](-0.4949,0,1,1)


(Z_critical =-1.645)

Zobs> Z_critical (p-value>0.05)


We do not reject the null hypothesis


At the 10% level of significance, we also fail to reject the null


hypothesis, leading to the same conclusion.


In summary, based on the sample data, we wouldn't conclude that


the fiber is unacceptable at both the 5% and 10% levels of

significance.

4. Assumption:

Random samples are drawn from the population

. H 0 : μ ≥5 vs H 1 : μ <5
Test Statistic : 7 (4.8-5)/0.5=-2.8

p-value =P(z<-2.8)=0.002555 ([Link](-2.8,0,1,1)



p-value<0.05
We reject the null hypothesis

5. Assumption: Random samples are drawn from the population.


H 0 : μ ≤15 vs H 1 : μ >15 ( Right Sided test)

Test statistic : zobs=10*(16-15)/2=5

p-value : P(Z>zobs)=0 ([Link](5,0,1,1)

p-value< 0.05, Reject Null hypothesis


6. a) Type -1 error, b) Type 2 Error

7 a) . H 0 : μ=50 vs H 1 : μ ≠ 30

b) p =proportion of the days stock price increases. . H 0 : p ≤ 0.6 vs


H 1 : p> 0.6

c) H 0 : μ ≥7 vs H 1 : μ <7

8. Test statistic: sqrt(16)(95-100)/10=-2

Decision : Reject

9. . H 0 : μ=150 vs H 1 : μ ≠ 150

Test statistic: sqrt(64)*(158-150)/20=3.2

Critical value approach :

Determine the critical value for a two-tailed test at α=0.05: Since it's a two-
tailed test with

α =0.05, the significance level is split into two tails. Therefore, each tail has
α/2=0.025.
The critical z-value corresponding to the left tail is z=−1.96
([Link](0.025,0,1)).

 The critical z-value corresponding to the right tail is z=1.96.

 From the earlier calculations, we found that our test statistic z=3.2

Our calculated z-statistic (3.2) is greater than the positive critical value
(1.96). Therefore, it lies in the rejection region of the right tail.

Given that our test statistic lies in the rejection region, we reject the null
hypothesis

p-value approach

zobs=3.2 |zobs|=3.2, 2*P(Z>|zobs|)= 0.0013 (2*([Link](3,2,0,1,1))

p-value <0.05, we reject the null hypothesis.

10. H 0 : μ ≤50 vs H 1 : μ >50, z-test, Reject.

11. H 0 : μ=7 vs H 1 : μ ≠ 50, t-test, Reject.

12 . H 0 : μ=24 7 vs H 1 : μ ≠ 247 , z-test (Also you can use t-test assuming normality),
Do not reject
13. H 0 : μ=102.5 vs H 1 : μ ≠ 107.5, t-test, Reject at 5%, Do not reject at 1%

17. Hypotheis H 0 : μ ≤.15 H 0 : μ>0.15

Alpha =0.05
Assumption : i) Samples are independent.

Mean=0.162, s.d=.040, n=40


Test statistic =: tobs= sqrt (40) *(0.162-0.15)/.040 =1.897367 , Reject H 0

18. Hypothess:
H 0 : p=0.05 vs H1 p≠0.05

n=200 np=200*0.05=10 n(1-p)=200*0.95=190, So conditions for normality are satisfied

sample proportion =14/200=0.07

Zobs = SQRT(200)* (0.07-0.05)/ SQRT(0.05*0.95)=1.297771


Critical Value Approach :
Zobs> [Link](0.975,0,1) or Zobs < [Link](0.025,0,1), we reject null hypothesis,
otherwise we do not reject.
[Link](0.975,0,1)=1.96, [Link](0.025,0,1) =-1.96
Zobs=1.297771 , so we do not reject.
p-value =2P(Z>|Zobs|)= 2*([Link](1.297771,0,1,1))= 0.194366>0.05, We do not
reject.

19.

H 0 : p ≤ 0.20 vs H1 p > 0.20

n=100 np=100*0.20=20 n(1-p)=100*0.80=80, So conditions for normality are satisfied

sample proportion =27/100=0.27

Zobs = SQRT(100)* (0.27-0.20)/ SQRT(0.20*0.80)= 1.75

p-value =P(Z>Zobs)=([Link](1.75,0,1,1))=0.040059<0.05, We reject null hypothesis.

20.

H 0 : p ≥ 0.60 vs H1 p < 0.60

Alpha=0.05

n=500 np=500*0.60= 300 n(1-p)=500*0.40= 200, So conditions for normality are satisfied

Sample proportion = 280/500= 0.56


Zobs = SQRT(500)* (0.56-0.60)/ SQRT(0.60*0.40)= -1.82

Z_critical =norm. [Link](0.05)= -1.64 We reject null hypothesis.

But if we set alpha=0.01, z_critical= [Link](0.01),=-2.32 we do not reject.

21. H 0 : μ=3000 vs H 1 : μ ≠ 3000 t-test,

22. H 0 : μ=50 vs H 1 : μ <50 t-test,

23 . H 0 : μ=15 vs H 1 : μ >15 t-test,

24 . H 0 : μ=8 vs H 1 : μ <8 z-test (t-test, assuming normality)

25 . H 0 : μ=75 vs H 1 : μ <75 z-test, (t-test, assuming normality)

26. H 0 : p = 0.60 vs H1 p≠ 0.60, z-test

27. H 0 : p ≥ 0.60 vs H1 p< 0.60, z-test

28. H 0 : p ≥0.70 vs H1 p<0.70, z-test

You might also like