Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
1K views

Distribution Test Statistic / Formula Conditions

This document provides an overview of hypothesis testing, including: - Hypothesis testing is used to assess the plausibility of a hypothesis using sample data. - Several common distributions and their test statistics/formulas are presented for performing hypothesis tests, including t, z, and proportions tests. - Examples of hypothesis tests are worked through step-by-step. - Practice problems involving hypothesis testing are presented at the end.

Uploaded by

Natalya Lewis
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
1K views

Distribution Test Statistic / Formula Conditions

This document provides an overview of hypothesis testing, including: - Hypothesis testing is used to assess the plausibility of a hypothesis using sample data. - Several common distributions and their test statistics/formulas are presented for performing hypothesis tests, including t, z, and proportions tests. - Examples of hypothesis tests are worked through step-by-step. - Practice problems involving hypothesis testing are presented at the end.

Uploaded by

Natalya Lewis
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

INSB 2015 – Applied Statistics for Business and Analytics

Tutor: Linda Deonath


Email: l.deonath@fac.gsb.tt

Tutorial 2
HYPOTHESIS TESTING

Hypothesis testing is used to assess the plausibility of a hypothesis by using sample data.
The test provides evidence concerning the plausibility of the hypothesis, given the data.
We test a hypothesis by measuring and examining a random sample of the population being
analyzed.
The following are formulae required to perform Hypothesis Testing.

Distribution Test Statistic / Formula Conditions


t- distribution 𝑥̅ − 𝜇0 n < 30
𝑡𝑐 = 𝑠
𝑥̅ = sample mean ⁄ 𝑛  (population standard
𝜇0 = population mean √
deviation) unknown
n – 1 degrees of freedom
s = sample standard deviation One population, single mean
n = sample size
Normal - distribution 𝑥̅ − 𝜇0 n < 30
𝑧𝑐 = 𝜎
𝑥̅ = sample mean ⁄ 𝑛  (population standard
√
𝜇0 = population mean deviation) known
 = population standard One population, single mean
deviation
n = sample size
Normal - distribution 𝑥̅ − 𝜇0 n > 30
𝑧𝑐 = 𝑠
𝑥̅ = sample mean ⁄ 𝑛  (population standard
𝜇0 = population mean √
deviation) unknown
s = sample standard deviation One population, single mean
n = sample size
Normal - distribution 𝑥̅ − 𝜇0 n > 30
𝑧𝑐 = 𝜎
𝑥̅ = sample mean ⁄ 𝑛  (population standard
√
𝜇0 = population mean deviation) known
 = population standard One population, single mean
deviation
n = sample size
Normal - distribution 𝑥̅1 − 𝑥̅2 𝑛1 , 𝑛2 > 30 or
𝑧𝑐 =
𝑥̅1 , 𝑥̅2 = sample means 𝑠1 , 𝑠2
𝑠1 , 𝑠2 = sample standard 𝑠12 𝑠22
√ Two independent populations,
𝑛1 + 𝑛2
deviations two means
𝑛1 , 𝑛2 = sample sizes
Normal - distribution 𝑝̂ − 𝑝0 𝑛𝑝̂ ≥ 10, 𝑛(1 − 𝑝̂ ) ≥ 10 or
𝑧𝑐 =
𝑝0 = population proportion
𝑝̂ = sample proportion √𝑝0 (1 − 𝑝0 )
𝑛
𝑛= sample size
Normal - distribution (𝑝̂1 − 𝑝̂ 2 ) − (𝑝1 − 𝑝2 ) 𝑥1 + 𝑥2
𝑧𝑐 = 𝑃̂ =
𝑝1 , 𝑝2 = population 1 1 𝑛1 + 𝑛2
proportions √𝑃̂ (1 − 𝑃̂) ( + )
𝑛1 𝑛2
𝑝̂1 , 𝑝̂ 2= sample proportions
𝑥1 , 𝑥2 = number of successes
in respective samples
𝑛1 , 𝑛2 = sample sizes
𝑃̂ = pooled proportions
Examples:
1. The average mass of chocolate bars produced at a candy factory is 200g. Tim, in quality
control, believes the mass to be different and that the machines need to be recalibrated.
He measured the mass of 100 chocolate bars and found the mean to be 201.5g with a
standard deviation of 6.5.
a) State the null and alternative hypotheses.
b) At a 95% confidence level, is there enough evidence to discard the null hypothesis?

H0:  = 200
H1:  ≠ 200
Since the mean is NOT EQUAL to a certain number, we have a two-tailed test. Given
information:
Sample size, n = 100
Sample mean,𝑥̅ = 201.5
Sample S.D. = 6.5
Confidence level = 95% or 0.95
Significance level ,  = 1 – 0.95 = 0.05

Let us first compare the calculated z-value with the z-value that is associated with the test
statistic.
The associated z-value is 1.96 as seen from picture above.

The z value is now calculated:


𝑥̅ −𝜇
𝑧𝑐 = 𝑆𝐷 0
⁄
√𝑛
201.5−200
𝑧𝑐 = 6.5⁄
√100
𝑧𝑐 = 2.31
So at a confidence level of 95%, since 2.31 our calculated z value is greater than the critical
z value of 1.96 (and falls in the shaded rejection region, we reject the null hypothesis.
We shall now look at the p-values.
p-value <  reject H0
p-value ≥  fail to reject H0

2. A factory manufactures cars with a warranty of 5 years on the engine and transmission.
An engineer believes that the engine or transmission will malfunction in less than 5 years.
He tests a sample of 40 cars and find the average time to be 4.8 years with a standard
deviation of 0.50.
a. State the null and alternative hypotheses.
b. At a 2% significance level, is there enough evidence to support the idea that the
warranty should be revised?
3. A business owner is in the process of deciding whether or not to invest in a new factory
that refines oil in order to meet the high demand for that commodity. A test showed that
the old factory refines oil at a mean rate of 3.1 L per second at a stand deviation of 1.0
using a sample size of 40. The new factory was measured to refine oil at a mean rate of
3.8 L per second at a standard deviation of 1.5 using a sample size of 36. Determine at a
10% significance level if there is any major difference.
4. A tech company believes that the percentage of residents in town XYZ that own a cell
phone is 70%. A marketing manager believes this value to be different. He conducts a
survey of 200 individuals and found that 130 responded yes to owning a cell phone.
State the null and alternative hypotheses and determine, at a 95% confidence level,
whether there is enough evidence to reject the null hypothesis.

5. A car company believes that the percentage of residents in city ABC that own a vehicle is
60% or less. A sales manager disagrees with this. He conducts a hypothesis test surveying
250 residents and found that 170 responded yes to owning a vehicle. State the null and
alternative hypotheses and determine at a 10% significance level, whether there is enough
evidence to support the idea that the vehicle ownership in city ABC is 60% or less?
6. Company XYZ manufactures laptops. For quality control, two sets of laptops were tested.
In the first group, 32 out of 800 were found to contain some sort of defect. In the second
group, 30 out of 500 were found to have a defect. Is the difference between the two
groups significant? (Use a 0.05 significance level).

Questions:

1. Given 𝑛 = 20, 𝑥̅ = 250, 𝜎 = 45, 𝛼 = 10%


𝐻0 : 𝜇 = 225
𝐻1 : 𝜇 ≠ 225
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0.

2. Given 𝑛 = 640, 𝑥̅ = 17500, 𝜎 = 2500, 𝛼 = 1%


𝐻0 : 𝜇 = 20000
𝐻1 : 𝜇 < 20000
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

3. Given 𝑛 = 1000, 𝑥̅ = 3500, 𝜎 = 4000, 𝛼 = 5%


𝐻0 : 𝜇 = 3000
𝐻1 : 𝜇 ≠ 3000
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

4. Given 𝑛 = 50, 𝑥̅ = 4.5, 𝜎 = 1.2, 𝛼 = 5%


𝐻0 : 𝜇 = 5
𝐻1 : 𝜇 ≠ 5
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

5. Given 𝑛1 = 200, 𝑥̅1 = 1.2, 𝜎1 = 0.07, 𝛼 = 5%


𝑛2 = 100, 𝑥̅2 = 1.15, 𝜎2 = 0.05
𝐻0 : 𝜇1 = 𝜇2
𝐻1 : 𝜇1 > 𝜇2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

6. Given 𝑛1 = 100, 𝑥̅1 = 10.49, 𝜎1 = 0.08, 𝛼 = 5%


𝑛2 = 80, 𝑥̅2 = 10.47, 𝜎2 = 0.05
𝐻0 : 𝜇1 = 𝜇2
𝐻1 : 𝜇1 ≠ 𝜇2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

7. Given 𝑛1 = 400, 𝑥̅1 = 125, 𝜎1 = 22, 𝛼 = 5%


𝑛2 = 300, 𝑥̅2 = 110, 𝜎2 = 27
𝐻0 : 𝜇1 = 𝜇2
𝐻1 : 𝜇1 ≠ 𝜇2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

8. Given 𝑛1 = 200, 𝑥̅1 = 250, 𝜎1 = 45, 𝛼 = 10%


𝑛2 = 150, 𝑥̅2 = 220, 𝜎2 = 55
𝐻0 : 𝜇1 = 𝜇2
𝐻1 : 𝜇1 ≠ 𝜇2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

9. Given 𝑛 = 500, 𝑥 = 412, 𝑝 = 0.90, 𝛼 = 5%


𝐻0 : 𝑝 = 0.90
𝐻1 : 𝑝 ≠ 0.90
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

10. Given 𝑛 = 1000, 𝑥 = 490, 𝑝 = 0.50, 𝛼 = 5%


𝐻0 : 𝑝 = 0.50
𝐻1 : 𝑝 > 0.50
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

11. Given 𝑛 = 1000, 𝑥 = 24, 𝑝 = 0.02, 𝛼 = 5%


𝐻0 : 𝑝 = 0.02
𝐻1 : 𝑝 < 0.02
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

12. Given 𝑛 = 1400, 𝑥 = 644, 𝑝 = 0.50, 𝛼 = 5%


𝐻0 : 𝑝 = 0.50
𝐻1 : 𝑝 > 0.50
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

13. Given 𝑛1 = 500, 𝑝1 = 0.25, 𝛼 = 5%


𝑛2 = 400, 𝑝2 = 0.3
𝐻0 : 𝑝1 = 𝑝2
𝐻1 : 𝑝1 ≠ 𝑝2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

14. Given 𝑛1 = 200, 𝑥1 = 70, 𝛼 = 5%


𝑛2 = 100, 𝑥2 = 32,
𝐻0 : 𝑝1 = 𝑝2
𝐻1 : 𝑝1 ≠ 𝑝2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

15. Given 𝑛1 = 500, 𝑥1 = 260, 𝛼 = 5%


𝑛2 = 400, 𝑥2 = 250,
𝐻0 : 𝑝1 = 𝑝2
𝐻1 : 𝑝1 < 𝑝2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

16. Given 𝑛1 = 1000, 𝑝1 = 0.25, 𝛼 = 5%


𝑛2 = 800, 𝑝2 = 0.2
𝐻0 : 𝑝1 = 𝑝2
𝐻1 : 𝑝1 ≠ 𝑝2
Determine whether we should reject H0 in favor of the alternative hypothesis, or fail to
reject H0

You might also like