Hypothesis Testing

© All Rights Reserved

1 views

Hypothesis Testing

© All Rights Reserved

- MQM100_MultipleChoice_Chapter9
- Stickleback Analysis
- Thabane L. Sample Size Determination in Clinical Trials. 2004.pdf
- Chapter 13 Solutions
- Sampling
- Hypothesis
- A 5
- MB0034
- SMU MBA QUESTION PAPER
- MB0050-RESEARCH METHODOLOGY (rev)
- ihe734_042611
- The Effect of Accounting Procedures Changes on Executive Remuneration
- Hypothesis Testing - 1
- day7hypothesistestingforposting-120903104108-phpapp02
- Erikson Et Al. - Dyadic Analysis in International Relations-A Cautionary Tale
- Comparison of Key Economic Indicators of Pakistani Economy
- Horizontal price transmission of the Finnish.pdf
- Logic Hypothesis
- Ej 1060320
- part 4.pdf

You are on page 1of 4

Objectives

Identify the differences between populations and samples

Define the terms related to hypothesis testing

Learn the algorithm for solving hypothesis testing problems for normally

distributed, variance-known data

Apply more complicated types of hypothesis testing, including tests on mean with

variance unknown and tests on variance

Our work thus far has been on populations, where we are considering an entire possible range

of data. Now, we will work on samples, which are manageable subsets of a population. So

before we would be considering population mean and standard deviation (µ and σ), and now

we are considering the sample mean and standard deviation (𝒙 ̅ and 𝒔).

Hypothesis Testing

In real life we will mostly be working with samples, but wanting to make inferences about

populations. This is where hypothesis testing comes in. Basically, we will create a hypothesis

about parameters of a population (generally means or variances) then use the sample

data to see if we can accept or reject the hypothesis.

The following are a list of terms having to do with hypothesis testing. Some of these terms are

not relevant to today’s lab but will be useful when we solve more complicated hypothesis testing

problems.

Null hypothesis (H0) – The statement we are testing. Generally, we will be testing

whether or not two things are equal to each other: for example, H0: µ = 100.

Alternative hypothesis (H1) – An opposite statement to the null. If the problem

statement suggests we are testing if a mean is exactly equal to a number, we use

the alternative hypothesis H1: µ ≠ 100. If the problem suggests we want to know if

a mean is less than or greater than a number, we use H1: µ < 100 or H1: µ > 100. H1

≠ µ is a two-sided hypothesis because are considering if the mean is greater than OR

less than a value, or to either side of the mean in the distribution. The other alternative

hypotheses are one-sided hypothesis because we are only considering one side of the

distribution.

Type I error – This occurs when we reject the null hypothesis, but it is actually true.

Significance level (α) – This is the probability of a type I error occurring, or the

probability of rejecting our null hypothesis even though it is true.

P-value – This is the minimum level of significance that would lead to rejecting the null

hypothesis. In other words, if the P-value is less than the significance level (α), we

will reject the null hypothesis.

Two-tailed test – We perform this test when our alternative hypothesis is two-sided. If it

is two-sided, you must consider HALF of the significance level or P-value on either end

of the distribution; for example, if your significance level is 0.5, or 5%, you divide that by

two to consider 2.5% of the data under the tail on the left and 2.5% of data under the tail

on the right.

One-tailed test – We perform this test when our alternative hypothesis is one-sided.

Type II error – This error occurs when we accept the null hypothesis (or fail to reject)

but it is actually false.

2. Normalize your data by converting your test statistic (whatever µ is compared to in your

hypotheses) into a Z value (if your variance is known – later we will consider if the

variance is unknown).

𝑥̅ − 𝜇

𝑍=

𝜎/√𝑛

3. Find the probability corresponding to this Z value using the charts in your book or Excel

functions.

4. If the P-value is less than your significance level, you can reject the null hypothesis.

Lab 5 Exercises

You want to test if the mean of your process data is equal to 100. You know that σ = 3 and n =

9.

1. State your null and alternative hypothesis. Is this a one-sided or two-sided test?

2. What is your P-value if 𝑥̅ = 103?

3. If the rejection region is defined as 𝑥̅ < 98.5 and 𝑥̅ > 101.5, what is α?

Now we want to test if the mean of the process data is less than 100 and we still know that σ =

3 and n = 9.

4. State your null and alternative hypothesis. Is this a one-sided or two-sided test?

5. What is your P-value if 𝑥̅ = 99?

Now we’ll expand to consider other tests we may do. Essentially, depending on what we are

testing and if the variance is known or unknown, we use a different statistic with a different

distribution, but the same algorithm that we learned last week to solve problems. For example,

for a test on the mean with variance known, we used the Z statistic. The various tests are

outlined below. For more information, refer to the section in the book.

Mean, variance known 4-4 𝑋̅ − 𝜇0 normsdist

𝑍= normsinv

𝜎/√𝑛

Mean, variance 4-5 𝑋̅ − 𝜇0 tdist(x,v,tails)

unknown 𝑇= tinv(p,v)

𝑆/√𝑛

x = standardized T,

using the equation to

the left

v = n-1, the degrees

of freedom

tails = 1 or 2,

depending on the

number of tails for

the alternative

hypothesis

p = probability

Variance 4-6 𝑆 2 (𝑛 − 1) chidist(x,v)

𝜒2 = chiinv(p,v)

𝜎02

x = standardized 𝜒 2 ,

using the equation to

the left

v = n-1, the degrees

of freedom

p = probability

Lab 6 Exercises

For 𝑋̅ = 61492, s = 3035, and n = 10:

1. State the appropriate hypothesis test to see if the sample mean is greater than 60000.

2. Find the P-value for this test.

3. Find the 90% upper confidence (one sided) interval bound on the mean (μ).

4. State the appropriate hypothesis test to see if the sample standard deviation is greater

than 0.02.

5. Find the P-value for this test.

6. Say the 95% confidence interval for the test is 0 ≤ 𝜎 ≤ 0.029. Does the boundary value

agree with what you expected from the P-value?

- MQM100_MultipleChoice_Chapter9Uploaded byNakin K
- Stickleback AnalysisUploaded byAlec Reder
- Thabane L. Sample Size Determination in Clinical Trials. 2004.pdfUploaded byGede Ariyana Kusuma
- Chapter 13 SolutionsUploaded byGreg
- SamplingUploaded byRKP2209
- HypothesisUploaded byJamesJunior
- A 5Uploaded byJoshua Cummings
- MB0034Uploaded bychitradeepd09
- SMU MBA QUESTION PAPERUploaded bySAMDAVIDCB
- MB0050-RESEARCH METHODOLOGY (rev)Uploaded bymrp1616
- ihe734_042611Uploaded bycecsdistancelab
- The Effect of Accounting Procedures Changes on Executive RemunerationUploaded byFridRachman
- Hypothesis Testing - 1Uploaded bykill_my_klone
- day7hypothesistestingforposting-120903104108-phpapp02Uploaded byKritika Jain
- Erikson Et Al. - Dyadic Analysis in International Relations-A Cautionary TaleUploaded bytzengyen
- Comparison of Key Economic Indicators of Pakistani EconomyUploaded bywaqasjaved869673
- Horizontal price transmission of the Finnish.pdfUploaded byMisael
- Logic HypothesisUploaded byjulien17_2000
- Ej 1060320Uploaded byehlayzaa
- part 4.pdfUploaded byKarolina Polska
- Biostatistics.pdfUploaded byAnonymous 3xcMImL4
- Maths 4 NotesUploaded byeya
- Some Notes AdvertUploaded byRamnath Peethambaran
- MT230_2017-18Uploaded byecd4282003
- 1788295757Uploaded byTrendkill Trendkill Trendkill
- Navidi_ch6Uploaded byIsmail TiGuint
- EstudianteUploaded byYen Fishër
- smUploaded bydhaiwat
- ch09-BLACK.pptUploaded byV Kaul
- Module 1 - Tests of Hypothesis for a Single SampleUploaded byMaria Andres

- 64 Interview QuestionsUploaded byshivakumar N
- 1-s2.0-S0010482519300526-mainUploaded byAnthony Park
- TBI Quantitative Data GenerationUploaded byAnthony Park
- Pasteurization Process Energy Optimization for a Milk Dairy Plant by Energy Audit ApproachUploaded byIJSTR Research Publication
- Spence, Jonathan D. - The Search for Modern China (1990)Uploaded byTian Yan
- Machiavelli, Niccolò - Discourses on Livy (2007)Uploaded byauto3871
- Diana Wynne Jones - The Tough Guide to Fantasy LandUploaded byDaniel Wakefield
- RulebookUploaded bymiguel barros
- Law Professor Cashin Makes Case for Building Affirmative Action AnewUploaded byAnthony Park
- Why Start With WhyUploaded byAnthony Park
- Reological%20Methods%20in%20Food%20Processing.pdfUploaded byuapaza
- Syllabus TLI521 Drug Development Spring 2019-APUploaded byAnthony Park
- The mechanism and function of circular RNAs in human diseasesUploaded byAnthony Park
- Final Deliverable ABE 301Uploaded byAnthony Park
- Requiem-Manual.pdfUploaded byJoão Machado
- Written Assignment #1 DraftUploaded byAnthony Park
- NISTperiodic TableUploaded byquelle1
- Machiavelli, Niccolò - Selected Political Writings (Hackett, 1994)Uploaded byTomislav Mijatov
- Economics - Adam Smith - The Wealth of Nations (Complete and Unabridged) - Mcgraw-HillUploaded bypbmbuss

- Welch tUploaded byChhavi Nagpal
- Research PptUploaded byAsif Saeed
- Aptitude TestsUploaded byGagan Singh
- Tips for Scoring 330+ on the GRE - CrunchPrep GREUploaded byRaúl Duarte González
- ReliabilityUploaded bymagicboy88
- Cover SkripsiUploaded byAudityaErosSiswanto
- Validity and ReliabilityUploaded bypalooo4
- Appedix AUploaded byCiv Nortub
- vef1Uploaded byanjauiah
- CMA+Exam+Secrets+Handout.pdfUploaded byInah Salcedo
- Repeated Measure DesignUploaded byChriselle R
- Paulhus - Abilitati CognitiveINTELLIGENCE 04 With HarmsUploaded byElena Diana
- Statistical Basics for Micro Array StudiesUploaded bypateldhaval.in
- multiple analysisUploaded byHazel Flores
- 2017 2018 texas university requirements 2Uploaded byapi-287225286
- dell soln 4 - 8Uploaded byAnkit Jani
- » OUTLOOK India Survey 2013 _ Top 75 Engineering Colleges in IndiaUploaded byBhanu Partap Sharma
- Module in Statistics and ProbabilityUploaded byMichael Knight
- QMT11 Chapter 9 Hypothesis TestingUploaded byBamPanggat
- Clerical Reasoning Test Technical ManualUploaded bym.rose
- List of BooksUploaded byIrfan Sayyed
- Mabini CollegesUploaded byRoderick Gatdula
- act vs sat styleUploaded byapi-241148422
- Koppitz-2- Reynolds.pptUploaded bymemo ley
- Hypothesis TestingUploaded byshomitamaraewu
- https___cdn4.digialm.com___per_g01_pub_585_touchstone_AssessmentQPHTMLMode1__GATE1764_GATE1764S3D10937_15179194948997247_CS18S35017299_GATE1764S3D10937E1Uploaded bykrishnapal singh tiwari
- bab 3 UCULUploaded byDesy Apridianty
- intro to hypothesis testingUploaded byrahulrockon
- 3. Hypothesis TestingUploaded bymonerch Joser
- APICS scoringUploaded byAnubhav Pandey

## Much more than documents.

Discover everything Scribd has to offer, including books and audiobooks from major publishers.

Cancel anytime.