• Home
  • About
    • Yerim Oh photo

      Yerim Oh

      Happy and worthwhile day by day :)

    • Learn More
    • Email
    • LinkedIn
    • Instagram
    • Github
    • Youtube
  • Posts
    • All Posts
    • All Tags
  • Projects

[12] STATISTICS: One-Sample Tests_one tail test

01 Mar 2020

Reading time ~3 minutes

Table of Contents
  • λͺ©μ°¨
  • Fundamentals of Hypothesis Testing: One-Sample Tests
  • κ°œμš”
  • INTRO
  • One-Tail Tests μ’…λ₯˜
    • Lower-Tail Tests
    • Upper-Tail Tests
  • Example: Upper-Tail t Test for Mean (𝜎 unknown)
    • PROBLEM
    • SOLVE 1
    • SOLVE 2: p-value

λͺ©μ°¨

  • Fundamentals of Hypothesis Testing: One-Sample Tests
  • κ°œμš”
  • INTRO
  • One-Tail Tests μ’…λ₯˜
    • Lower-Tail Tests
    • Upper-Tail Tests
  • Example: Upper-Tail t Test for Mean (𝜎 unknown)
    • PROBLEM
    • SOLVE 1
    • SOLVE 2: p-value

πŸ‘€, πŸ€·β€β™€οΈ , πŸ“œ, πŸ“
이 μ•„μ΄μ½˜λ“€μ„ λˆ„λ₯΄μ‹œλ©΄ μ •λ‹΅, κ°œλ… λΆ€κ°€ μ„€λͺ…을 보싀 수 μžˆμŠ΅λ‹ˆλ‹€:)



Fundamentals of Hypothesis Testing: One-Sample Tests

즉, 가섀검증이닀.
testλΌλŠ” μœ„λ”©μ΄ λ“€μ–΄κ°”λ‹€λŠ” 것은 이 μΆ”μ •μ˜ 결과둜 μΈν•œ μ˜μ‚¬κ²°μ •κΉŒμ§€ ν•œλ‹€λŠ” 것이닀.

One-Sample Tests: 데이터 μ„ΈνŠΈκ°€ ν•˜λ‚˜μ΄λ‹€.

  • One-Sample Tests: μš°λ¦¬λ‚˜λΌ μ‚¬λžŒμ˜ 아이큐가 100 이상이닀 -> μš°λ¦¬λ‚˜λΌ μ‚¬λžŒμ˜ μ•„μ΄νλΌλŠ” ν•˜λ‚˜μ˜ 데이터 μ„ΈνŠΈκ°€ ν•„μš”ν•˜λ‹€
  • Two-Sample Tests: μš°λ¦¬λ‚˜λΌμ™€ 일본의 아이큐λ₯Ό 비ꡐ해보면 μš°λ¦¬λ‚˜λΌκ°€ 더 λ†’λ‹€ -> μš°λ¦¬λ‚˜λΌ 아이쿠 데이터 μ„ΈνŠΈμ™€ 일본의 아이큐 λ°μ΄ν„°μ„ΈνŠΈκ°€ ν•„μš”ν•˜λ‹€. 즉, 비ꡐ, λ³€ν™” 검증에 많이 쓰인닀.

κ°œμš”

μ•„λž˜μ˜ 도식을 ν•˜λ‚˜μ”© μžμ„Ένžˆ μ•Œμ•„λ³Ό 것이닀.
image

  • Οƒ known: Fundamentals of Hypothesis Testing Methodology
  • Οƒ Unknown: t-Test of Hypothesis for the Mean
  • One-Tail Tests(도식 전체에 적용)
  • Population Proportion: Z-Test of Hypothesis for the Proportion
  • Potential Hypothesis Testing Pitfalls and Ethical Issues(μΆ”κ°€)

INTRO

[One-Tail Tests(도식 전체에 적용)]
image

  • two-tail test: μ§€κΈˆκΉŒμ§€ λ°°μš΄κ²ƒμœΌλ‘œ 이게 λ§žλƒ μ•„λ‹ˆλƒμ˜ 문제
    image
  • one-tail test: 이거보닀 μž‘λƒ 이거보닀 ν¬λƒμ˜ 문제
    image


One-Tail Tests μ’…λ₯˜

  • Lower-Tail Tests image
  • Upper-Tail Tests image

Lower-Tail Tests

image

[κ°€μ„€μ˜ λ°©ν–₯]
alternative hypothesis \(H_1\) is focused on the lower tail
\(H_0\), \(H_1\) 더 μ•Œμ•„λ³΄κΈ°
(EX)

  • \(H_0 : ΞΌ\) \(>=\) \(3\) ➑️ = 이 λ“€μ–΄κ°€λŠ” μ΄μœ λŠ” ν˜„μž¬μ˜ 것을 acceptν•˜κΈ° λ•Œλ¬Έμ΄λ‹€.
  • \(H_1 : ΞΌ < 3\).

[critical value]
There is only one critical value, since the rejection area is in only one tail
λ˜ν•œ a값이 a/2κ°’ (two-tail test) 이 μ•„λ‹Œ aκ·Έ 자체 이닀.


Upper-Tail Tests

image

[κ°€μ„€μ˜ λ°©ν–₯]
alternative hypothesis \(H_1\) is focused on the upper tail
\(H_0\), \(H_1\) 더 μ•Œμ•„λ³΄κΈ°
(EX)

  • \(H_0 : ΞΌ\) \(<=\) \(3\)
  • \(H_1 : ΞΌ > 3\).

[critical value]
There is only one critical value, since the rejection area is in only one tail
λ˜ν•œ a값이 a/2κ°’ (two-tail test) 이 μ•„λ‹Œ aκ·Έ 자체 이닀.



Example: Upper-Tail t Test for Mean (𝜎 unknown)

그럼 이제 μœ„μ˜ κ°œλ…μ„ μ΄μš©ν•΄ 문제λ₯Ό 풀어보며 이해λ₯Ό ν•΄λ³΄μž!
그럼 무슨 κ°œλ…μΈμ§€ 더 와닿을 것이닀.

λ¨Όμ € λ¬Έμ œμ— μΉœμˆ™ν•΄μ§€κΈ° μœ„ν•΄ 𝜎 unknown의 two-tail 버전을 λ³΅μŠ΅ν•˜κ³  였자!


PROBLEM

A phone industry manager thinks that customer monthly cell phone bills have increased, and now average over $52 per month. The company wishes to test this claim. (Assume a normal population)

➑️ now average over $52 per month: 52λ³΄λ‹€λŠ” λ§Žλ‹€ μ΄λ―€λ‘œ one-tail μ΄λΌλŠ” 것을 μ•Œ 수 μžˆλ‹€.

[Form hypothesis test:]

  • \(H_0 : ΞΌ\) \(<=\) \(52\)
  • the mean is not over $52 per month
  • \(H_1 : ΞΌ > 52\).
  • the mean is greater than $52 per month

SOLVE 1

1) Find Rejection Region

  • Suppose that a = 0.10 is chosen for this test and n = 25.
  • Find the rejection region:
    image

2) Test Statistic
Obtain sample and compute the test statistic
Suppose a sample is taken with the following results:

  • n = 25, = 53.1, and S = 10
  • Then the test statistic is
    image

3) Decision
Reach a decision and interpret the result:
image ➑️ Do not reject \(H_0\) since \(t_{STAT}\) = 0.55 < 1.318

  • there is not sufficient evidence that the mean bill is over $52

SOLVE 2: p-value

p-value λž€?

Calculate the p-value and compare to a
image

➑️ Do not reject \(H_0\) since p-value = .2937 > a = .10

  • there is not sufficient evidence that the mean bill is over $52


Mathematics Share Tweet +1