Two Samples Estimation: Paired Observations
βνλ₯ κ³Ό ν΅κ³(MATH230)β μμ μμ λ°°μ΄ κ²κ³Ό 곡λΆν κ²μ μ 리ν ν¬μ€νΈμ λλ€. μ 체 ν¬μ€νΈλ Probability and Statisticsμμ νμΈνμ€ μ μμ΅λλ€ π²
Interval Estimation ν¬μ€νΈμμ λ€λ£¬ <Interval Estimation>μ νΉμ μν©μ μ΄λ»κ² μ μ©ν μ μλμ§λ₯Ό λ€λ£¨λ ν¬μ€νΈμ λλ€.
μλμ λ¬Έμ λ₯Ό μ΄ν΄λ³΄μ!
μ°λ¦¬λ νμ1λΆν° νμ30κΉμ§ κ·Έλ€μ TOEIC μ μμ before-afterλ₯Ό κ°μ§κ³ μλ€. μ°λ¦¬λ κ³Όμ° MATH230 μμ μ΄ νμλ€μ TOEIC μμ μ μ΄λ€ μν₯μ λ―ΈμΉλμ§ μκΈ° μν΄ $\mu_1 - \mu_2$λ₯Ό μΆμ νκ³ μ νλ€!!
Q. Can we find a 95% confidence interval for the true mean of the differences btw the scores before and after the MATH230?
Supp. $X_1, \dots, X_n$ and $Y_1, \dots, Y_n$ are random samples and $\sigma_1^2$ and $\sigma_2^2$ are known.
μ΄μ ν¬μ€νΈ βTwo Samples Estimation: Diff Btw Two Meansβμμ λ§μ½ λ μνμ λΆμ°μ μ νν μλ€λ©΄, μλμ κ°μ΄ ꡬκ°μ μΆμ ν μ μλ€κ³ νμλ€.
\[\left| \bar{x} - \bar{y} \right| \le z_{\alpha/2} \cdot \sqrt{\frac{\sigma_1^2}{n} + \frac{\sigma_2^2}{n}}\]π₯ ν!μ§!λ§! μμ λ°©λ²μ μ¬λ°λ₯Έ μ κ·Όμ΄ μλλ€! μλνλ©΄, νμ¬ μ°λ¦¬κ° κ°μ§ μν $X_i$, $Y_i$μ λν΄ κ·Έ λμ΄ μλ‘ dependent νκΈ° λλ¬Έμ΄λ€!! μμ μ κ·Όμ $X_i$μ $Y_i$κ° independent ν λλ§ κ°λ₯νλ€!!
κ·Έλμ μ°λ¦¬λ κ° $X_i$, $Y_i$λ₯Ό κ°λ³μ μΌλ‘ μκ°νλ κ²μ΄ μλλΌ κ·Έλ€μ paringν Difference $D_i = X_i - Y_i$λ‘ μ κ·Όνκ³ μ νλ€!
μ΄λ κ² ν κ²½μ°, κ° Pairλ μλ‘ independentνκ² λλ€!
Assume that $D_1, \dots, D_n$ are normal random samples: $D_i \sim N(\mu_D, \sigma_D^2)$
To find the confidence interval for $\mu_1 - \mu_2$, we use $\bar{D} := \bar{X} - \bar{Y}$.
Then, by CLT
\[\frac{\bar{D} - \mu_D}{\sigma_D / \sqrt{n}} \; \sim \; N(0, 1)\]μ΄λ, μ°λ¦¬λ $\sigma_D^2$λ₯Ό μμ§ λͺ»νλ―λ‘ μ΄κ²μ sample variance $s_D^2$μΌλ‘ κ΅μ²΄νλ©΄ λΆν¬λ μλμ κ°λ€.
\[\frac{\bar{D} - \mu_D}{s_D / \sqrt{n}} \; \sim \; t(n-1)\]μ§κΈκΉμ§λ <Normal Distribution>μμ λ½μ random sampleμμ μΆμ (Estimation)μ μ§ννλ€. λ€μ ν¬μ€νΈμμλ <Bernoulli Distribution>μμ μννλ μΆμ μΈ <Proportion Estimation>μ λν΄ μ΄ν΄λ³Έλ€!! (Binomial Distributionμμμ νκ· μ Proportionμ΄λ€!! π)