β€œν™•λ₯ κ³Ό 톡계(MATH230)” μˆ˜μ—…μ—μ„œ 배운 것과 κ³΅λΆ€ν•œ 것을 μ •λ¦¬ν•œ ν¬μŠ€νŠΈμž…λ‹ˆλ‹€. 전체 ν¬μŠ€νŠΈλŠ” Probability and Statisticsμ—μ„œ ν™•μΈν•˜μ‹€ 수 μžˆμŠ΅λ‹ˆλ‹€ 🎲

3 minute read

β€œν™•λ₯ κ³Ό 톡계(MATH230)” μˆ˜μ—…μ—μ„œ 배운 것과 κ³΅λΆ€ν•œ 것을 μ •λ¦¬ν•œ ν¬μŠ€νŠΈμž…λ‹ˆλ‹€. 전체 ν¬μŠ€νŠΈλŠ” Probability and Statisticsμ—μ„œ ν™•μΈν•˜μ‹€ 수 μžˆμŠ΅λ‹ˆλ‹€ 🎲

Sample Mean Test

평균(Mean)에 λŒ€ν•œ 검정은 μΆ”μ •μ—μ„œμ™€ λΉ„μŠ·ν•˜κ²Œ, $\sigma^2$을 μ•„λŠ”μ§€ 여뢀에 따라 λ‹€λ₯΄κ²Œ μ ‘κ·Όν•œλ‹€.


$\sigma^2$ is known

1. 상황 인식

  • $H_0: \mu=190$
  • $H_1: \mu > 190$

$n=25$ and $\bar{x}=194$, and $\sigma^2 = 100$

2. Find a <Test Statistic>, and construct critical region

  • Test Statistic: $\bar{x}$
  • critical region: $\{ \bar{X} > C\}$

3. $\alpha$κ°€ 주어지지 μ•Šμ•˜μœΌλ‹ˆ, p-valueλ₯Ό κ΅¬ν•˜μž!

\[\begin{aligned} \alpha &= P(\bar{X} \ge 194 \mid \mu = 190) \\ &= P \left(\frac{\bar{X} - 190}{\sigma / \sqrt{n}} \ge \frac{194 - 190}{\sigma / \sqrt{n}} \right) \\ &= P(Z \ge 2) = 0.023 \end{aligned}\]

4. κ²°μ •

  • If $\alpha > 0.023$, reject $H_0$
  • If $\alpha < 0.023$, fail to reject $H_0$

$H_1: \mu < \mu_0$인 것도, $H_1: \mu \ne \mu_0$ κ²½μš°λ„ λΉ„μŠ·ν•˜κ²Œ, 식을 잘 μ„Έμ›Œμ„œ μ§„ν–‰ν•˜λ©΄ λœλ‹€!

μ •λ¦¬ν•˜λ©΄ μ•„λž˜μ™€ κ°™λ‹€.

Quick Remark.

사싀, β€œrejection region”과 β€œconfidence intervalβ€μ˜ μ„œλ‘œ Complementν•˜λ‹€!!

λ§Œμ•½, μš°λ¦¬κ°€ 얻은 sample mean $\bar{x}$이 $H_0$μ—μ„œ κ°€μ •ν•œ $\mu$의 confidence interval에 ν¬ν•¨λœλ‹€λ©΄,

\[\bar{x} \in (\mu \pm z_{\alpha/2} \cdot \sigma/\sqrt{n}) \quad \text{or} \quad \bar{x} < \mu + z_{\alpha} \cdot \sigma/\sqrt{n} \quad \text{or} \quad \bar{x} > \mu - z_{\alpha} \cdot \sigma/\sqrt{n}\]

μš°λ¦¬λŠ” $H_0$을 기각할 μ΄μœ κ°€ μ—†λ‹€. ν•˜μ§€λ§Œ, λ§Œμ•½ $\bar{x}$κ°€ confidence interval을 λ²—μ–΄λ‚œλ‹€λ©΄, μš°λ¦¬λŠ” μš°λ¦¬κ°€ μ„€μ •ν•œ $\mu$ 값을 μ˜μ‹¬ν•˜κ³ , 또 기각해야 ν•œλ‹€.

이것은 증λͺ… 방식 쀑 ν•˜λ‚˜μΈ β€œκ·€λ₯˜λ²•β€κ³Ό μœ μ‚¬ν•œλ°, β€œν†΅κ²Œμ • κ²€μ •(Testing)”은 β€œν™•λ₯ β€μ„ μ‚¬μš©ν•΄ 처음의 κ°€μ • $H_0$λ₯Ό κΈ°κ°ν•œλ‹€κ³  λ³Ό 수 μžˆλ‹€!


$\sigma^2$ is unknown

λ§Œμ•½, $\sigma^2$λ₯Ό λͺ¨λ₯Έλ‹€λ©΄, μΆ”μ •μ—μ„œ ν–ˆλ˜ κ²ƒμ²˜λŸΌ <t-test>λ₯Ό μ§„ν–‰ν•˜λ©΄ λœλ‹€. λ„ˆλ¬΄ μ‰¬μš°λ‹ˆ μ„€λͺ…은 생-랡 ν•˜κ² λ‹€.

μ •λ¦¬ν•˜λ©΄ μ•„λž˜μ™€ κ°™λ‹€.

πŸ’₯ 주의!! μƒ˜ν”Œμ€ 반.λ“œ.μ‹œ. Normal Distributionμ—μ„œ μΆ”μΆœλ˜μ–΄μ•Ό ν•œλ‹€!!


Two Samples Mean Test

이것도 사싀 별거 μ—†λ‹€. κ·Έλƒ₯ μΆ”μ •ν•΄μ„œ ν–ˆλ˜ 것과 μ•žμ—μ„œ ν–ˆλ˜ 것을 잘 λ…Ήμ—¬μ„œ 검정을 μˆ˜ν–‰ν•˜λ©΄ λœλ‹€.

μ •λ¦¬ν•˜λ©΄ μ•„λž˜μ™€ κ°™λ‹€.


λ§ˆμ°¬κ°€μ§€ λ°©λ²•μœΌλ‘œ <Test for Paired Observations>μ—μ„œλ„ κ·Έ.λŒ€.둜. 잘 μˆ˜ν–‰ν•˜λ©΄ λœλ‹€ 😁



λ‹€μŒ ν¬μŠ€νŠΈμ—μ„œλŠ” <κ²€μ •λ ₯; power of test> $\beta$λ₯Ό λ„μž…ν•΄ 평균(Mean)에 λŒ€ν•΄ κ²€μ •(Testing)을 μˆ˜ν–‰ν•  λ•Œ ν•„μš”ν•œ Sample Size $n$을 κ²°μ •ν•˜λŠ” 방법에 λŒ€ν•΄ μ‚΄νŽ΄λ³Έλ‹€.

πŸ‘‰ Choice of Sample Size for Testing Mean


이번 ν¬μŠ€νŠΈμ—μ„œλŠ” 평균(Mean)에 λŒ€ν•œ κ²€μ • 방법에 λŒ€ν•΄ μ‚΄νŽ΄λ΄€λ‹€. κ·Έλ ‡κ²Œ 어렡지 μ•Šμ•˜κ³ , μΆ”μ •(Estimation)μ—μ„œ ν•˜λ˜ κ±Έ, μ ˆμ°¨μ— 맞게 μˆ˜ν–‰ν•΄ 해석해주면 λ˜λŠ” κ±°μ˜€λ‹€. μ΄μ–΄μ§€λŠ” ν¬μŠ€νŠΈμ—μ„œλŠ” λΉ„μœ¨(proportion)κ³Ό λΆ„μ‚°(variance)에 λŒ€ν•œ 검정을 μ‚΄νŽ΄λ³Έλ‹€!

πŸ‘‰ Test on Proportion and Variance