โ€œํ™•๋ฅ ๊ณผ ํ†ต๊ณ„(MATH230)โ€ ์ˆ˜์—…์—์„œ ๋ฐฐ์šด ๊ฒƒ๊ณผ ๊ณต๋ถ€ํ•œ ๊ฒƒ์„ ์ •๋ฆฌํ•œ ํฌ์ŠคํŠธ์ž…๋‹ˆ๋‹ค. ์ „์ฒด ํฌ์ŠคํŠธ๋Š” Probability and Statistics์—์„œ ํ™•์ธํ•˜์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค ๐ŸŽฒ

11 minute read

โ€œํ™•๋ฅ ๊ณผ ํ†ต๊ณ„(MATH230)โ€ ์ˆ˜์—…์—์„œ ๋ฐฐ์šด ๊ฒƒ๊ณผ ๊ณต๋ถ€ํ•œ ๊ฒƒ์„ ์ •๋ฆฌํ•œ ํฌ์ŠคํŠธ์ž…๋‹ˆ๋‹ค. ์ „์ฒด ํฌ์ŠคํŠธ๋Š” Probability and Statistics์—์„œ ํ™•์ธํ•˜์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค ๐ŸŽฒ

Sampling Distribution of Mean

Let $X_1, \dots, X_n$ be a random sample with $E[X_i] = \mu$ and $\text{Var}(X_i) = \sigma^2$.

Then,

  • $E[\overline{X}] = \mu$
  • $\text{Var}(\overline{X}) = E\left[\left(\overline{X} - E[\overline{X}]\right)^2 \right] = \dfrac{\sigma^2}{n}$

<LLN; Law of Large Numbers>์— ๋”ฐ๋ฅด๋ฉด, $n$์ด ๋ฌดํ•œ์œผ๋กœ ๊ฐˆ๋•Œ, ๋ถ„์‚ฐ $\text{Var}(\overline{X}) = \sigma^2/n$๊ฐ€ 0์œผ๋กœ ์ˆ˜๋ ดํ•œ๋‹ค. ๋”ฐ๋ผ์„œ $\overline{X} \rightarrow \mu$๊ฐ€ ๋œ๋‹ค!


Weak Law of Large Numbers

Theorem. WLLN

Let $X_1, \dots, X_n$ be a random sample with $E[X_i] = \mu$ and $\text{Var}(X_i) = \sigma^2$.

Let $\overline{X}$ be a sample mean.

For any $\epsilon > 0$, we have

\[\lim_{n\rightarrow\infty} P\left(\left| \overline{X} - \mu \right| > \epsilon\right) = 0\]

Proof.

<Chebyshevโ€™s Inequality>๋ฅผ ์‚ฌ์šฉํ•˜๋ฉด ์•„์ฃผ ์‰ฝ๊ฒŒ ์ฆ๋ช…ํ•  ์ˆ˜ ์žˆ๋‹ค!

\[\begin{aligned} P\left(\left| \overline{X} - \mu \right| > \epsilon\right) &\le \frac{\text{Var}(\overline{X})}{\epsilon^2} \\ &= \frac{1}{\epsilon^2} \cdot \frac{\sigma^2}{n} \rightarrow 0 \quad \text{as} \quad n \rightarrow \infty \end{aligned}\]

$\blacksquare$

โ€œWLLN says that as the sample size $n$ gets larger, then the sample mean is close to the true mean in probability!โ€

์ด๋•Œ, WLLN๊ณผ ๊ฐ™์€ ํ˜•ํƒœ์˜ ์ˆ˜๋ ด์„ โ€œthe convergence in probabilityโ€๋ผ๊ณ  ํ•œ๋‹ค.

cf) ์ฐธ๊ณ ๋กœ <Strong Law of Large Numbers>๋„ ์กด์žฌํ•œ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์ด ์ •๋ฆฌ๋ฅผ ์ฆ๋ช…ํ•˜๋ ค๋ฉด, ์ธก๋„(measure)์— ๋Œ€ํ•œ ๊ฐœ๋…์ด ํ•„์š”ํ•˜๊ธฐ ๋•Œ๋ฌธ์— ์†Œ๊ฐœ๋งŒ ํ•˜๊ณ  ๋„˜์–ด๊ฐ€๊ฒ ๋‹ค.

\[P\left(\lim_{n\rightarrow\infty} \overline{X} = \mu \right) = 1\]

CLT; Central Limit Theorem

Example.

Q. What is the probability of $P\left( \overline{X} > 7\right)$?

Note that $X_1, \dots, X_n \sim \text{Ber}(p)$, and then $(X_1 + \cdots + X_n) \sim \text{BIN}(n, p)$.

When $n$ is largest, then $(X_1 + \cdots + X_n) \rightarrow N(\mu, \sigma^2)$.

Letโ€™s standardize it, then

\[P\left( \frac{(X_1 + \cdots + X_n) - np}{\sqrt{npq}} \le z \right) \approx P(Z \le z)\]

์ด๋•Œ, ์ขŒ๋ณ€์˜ ๋ถ„๋ชจ/๋ถ„์ž์— $n$๋ฅผ ๋‚˜์ค˜์ฃผ๋ฉด

\[\begin{aligned} P\left( \frac{((X_1 + \cdots + X_n) - np)/n}{(\sqrt{npq})/n} \le z \right) &= P\left( \frac{(X_1 + \cdots + X_n)/n \; - \; p}{\sqrt{pq/n}} \le z \right) \\ &= P\left( \frac{\overline{X} - E[\overline{X}]}{\sqrt{\text{Var}(\overline{X})}} \le z \right) \\ &\approx P(Z \le z) \end{aligned}\]

๊ฒฐ๋ก ์€, ์›๋ž˜ ๋ฌธ์ œ์˜€๋˜ $P\left(\overline{X} > 7\right)$์„ ์ž˜ ์ •๊ทœํ™”ํ•ด์„œ normal ๋ถ„ํฌ๋กœ ๊ทผ์‚ฌํ•˜์—ฌ ํ’€๋ฉด ๋œ๋‹ค๋Š” ๊ฒƒ์ด๋‹ค.

๊ทธ๋Ÿฐ๋ฐ, ์ง€๊ธˆ์˜ ๊ฒฝ์šฐ๋Š” $\overline{X}$๊ฐ€ BIN ๋ถ„ํฌ์˜€๊ธฐ ๋•Œ๋ฌธ์— <Normal Approximation>์— ์˜ํ•ด ์ž์—ฐ์Šค๋Ÿฝ๊ฒŒ ์œ ๋„๋œ ๊ฒƒ์ด์—ˆ๋‹ค. ๊ณผ์—ฐ $\overline{X}$ ๋˜๋Š” $X_i$๊ฐ€ ๋‹ค๋ฅธ ๋ถ„ํฌ๋ฅผ ๊ฐ€์ ธ๋„ ์œ„์™€ ๊ฐ™์€ ๋ฐฉ์‹์„ ํ’€ ์ˆ˜ ์žˆ์„๊นŒ? ์ด ์˜๋ฌธ์— ๋Œ€ํ•œ ๋‹ต์„ ์ œ์‹œํ•˜๋Š” ๊ฒƒ์ด ๋ฐ”๋กœ <CLT; Central Limit Theorem>์ด๋‹ค ๐Ÿคฉ


Theorem. CLT; Central Limit Theorem

Let $X_1, \dots, X_n$ be a random sample with $E[X_i] = \mu$ and $\text{Var}(X_i) = \sigma^2$.

Let $\overline{X}_n := (X_1 + \cdots + X_n)/n$, sample mean.

Let $Z_n := \dfrac{\overline{X}_n - E[\overline{X}]}{\sqrt{\text{Var}(\overline{X})}} = \dfrac{\overline{X}_n - \mu}{\sigma/\sqrt{n}}$.

then, for any $z \in \mathbb{R}$, we have

\[P(Z_n \le z) \rightarrow P(Z \le z) \quad \text{as} \quad n \rightarrow \infty\]

where $Z \sim N(0, 1)$.

์ฆ‰, ๋ชจ์ง‘๋‹จ์—์„œ ์ถ”์ถœํ•œ ํ‘œ๋ณธ $n$์ด ์ถฉ๋ถ„ํžˆ ํฌ๋‹ค๋ฉด, โ€œํ‘œ๋ณธํ‰๊ท โ€ $\bar{X}$์˜ ๋ถ„ํฌ๋Š” ์ •๊ทœ ๋ถ„ํฌ์— ๊ทผ์‚ฌํ•œ๋‹ค!

Remark.

1. As long as iid, RVs have finite second moment[^1], then we have CLT.

์ด๊ฒƒ์ด ์˜๋ฏธํ•˜๋Š” ๋ฐ”๋Š” ์•„์ฃผ ๊ฐ•๋ ฅํ•˜๋‹ค๐Ÿ’ฅ $X_i$๊ฐ€ ์–ด๋–ค ๋ถ„ํฌ๋ฅผ ๋”ฐ๋ฅด๋Š” ์ƒ๊ด€์—†์ด CLT๋ฅผ ์ ์šฉํ•  ์ˆ˜ ์žˆ๋‹ค๋Š” ๋ง์ด๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค!! ์ด๋Ÿฐ ์  ๋•Œ๋ฌธ์— CLT๋ฅผ โ€œuniversal resultโ€๋ผ๊ณ  ํ•œ๋‹ค!


2. We call $z: = \dfrac{\overline{x} - \mu}{\sigma / \sqrt{n}}$ as <$z$-value> or <$z$-score; $z$-์ ์ˆ˜, ํ‘œ์ค€ ์ ์ˆ˜>, we define $z_\alpha$ as the number $z$ s.t. $P(Z \ge z) = \alpha$ when $Z \sim N(0, 1)$.

Proof of CLT

Proof. CLT

<CLT>๋ฅผ ์ฆ๋ช…ํ•˜๊ธฐ ์œ„ํ•ด ์•„๋ž˜์˜ ์ •๋ฆฌ๋ฅผ ์‚ฌ์šฉํ•œ๋‹ค.

Theorem. Uniqueness of MGF

If the mgf of $X_n$ converges to the mgf of $X$ for $t \in (-\delta, \delta)$ for some $\delta > 0$,

i.e.

\[M_{X_n} (t) \rightarrow M_{X} (t) \quad \text{for} \quad t \in (-\delta, \delta)\]

and $X$ is continuous, then CDF $F_{X_n}(x)$ converges to $F_{X}(x)$ for all $x \in \mathbb{R}$.

\[F_{X_n}(x) \rightarrow F_{X}(x)\]

โœจ Goal: Show that the MGF of $Z = \dfrac{\bar{X} - \mu}{\sigma / \sqrt{n}}$ converges to the MGF of $N(0, 1)$ as $n \rightarrow \infty$.

Let $W = \dfrac{\bar{X} - \mu}{\sigma / \sqrt{n}}$, and then multiply $n$ both to the numerator and denominator.

\[W = \frac{\bar{X} - \mu}{\sigma / \sqrt{n}} = \frac{\displaystyle \sum_{i=1}^n X_i - n \mu}{\sqrt{n} \cdot \sigma}\]

The mgf of $W$ is

\[\begin{aligned} M_W (t) &= E [e^{tW}] = E\left[\exp \left( \frac{t}{\sqrt{n} \sigma}\sum^n_{i=1} X_i - n\mu \right) \right] \\ &= E \left[ \exp \left( \frac{t}{\sqrt{n}} \cdot \frac{X_1 - \mu}{\sigma}\right) \cdot \exp \left( \frac{t}{\sqrt{n}} \cdot \frac{X_2 - \mu}{\sigma}\right) \cdots \exp \left( \frac{t}{\sqrt{n}} \cdot \frac{X_n - \mu}{\sigma}\right) \right] \\ &= E \left[ \exp \left( \frac{t}{\sqrt{n}} \cdot \frac{X_1 - \mu}{\sigma}\right) \right] \cdots E \left[ \exp \left( \frac{t}{\sqrt{n}} \cdot \frac{X_n - \mu}{\sigma}\right) \right] \quad \text{iid} \\ &= M_{Z_1} (t / \sqrt{n}) \cdots M_{Z_n} (t / \sqrt{n}) \\ &= \left[ M_{Z} (t / \sqrt{n}) \right]^n \end{aligned}\]

์ด์ œ $M_W(t)$์— $\log$๋ฅผ ์ทจํ•˜๊ณ , ๊ทนํ•œ $n\rightarrow \infty$๋ฅผ ์ทจํ•˜๋ฉด,

\[\begin{aligned} \lim_{n\rightarrow \infty} \log M_W(t) \\ = \lim_{n\rightarrow \infty} \log \left[ M_{Z} (t / \sqrt{n}) \right]^n \\ &= \lim_{n\rightarrow \infty} n \cdot \log M_Z (t / \sqrt{n}) \end{aligned}\]

์—ฌ๊ธฐ์„œ $y = 1 / \sqrt{n}$๋กœ ์น˜ํ™˜ํ•ด์ฃผ๋ฉด, ์œ„์˜ ๊ทนํ•œ์‹์€ ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\lim_{n\rightarrow \infty} n \cdot \log M_Z (t / \sqrt{n}) = \lim_{y \rightarrow 0} \frac{\log M_Z (yt)}{y^2}\]

์ด๋•Œ, $\displaystyle \lim_{y\rightarrow 0} M_Z(yt) = M_Z(0) = 1$์ด๋ฏ€๋กœ ๊ทนํ•œ์‹์ด $\dfrac{0}{0}$ ๊ผด์˜ ๋ถ€์ •ํ˜•์ด ๋œ๋‹ค. ์ด๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด โ€œ๋กœํ”ผํƒˆ ์ •๋ฆฌโ€๋ฅผ ์‚ฌ์šฉํ•œ๋‹ค.

\[\begin{aligned} \lim_{y \rightarrow 0} \frac{\log M_Z (yt)}{y^2} &= \lim_{y \rightarrow 0} \frac{t \cdot \dfrac{M_z(yt)'}{M_Z(yt)}}{2y} \\ &= \frac{t}{2} \cdot \lim_{y \rightarrow 0} \frac{M_z(yt)'}{y \cdot M_Z (yt)} \\ &= \frac{t}{2} \cdot \lim_{y \rightarrow 0} \frac{M_z(yt)'}{y} \quad \left(\because \; \lim_{y\rightarrow 0} M_z(yt) = 1\right) \end{aligned}\]

์ด๋•Œ, ์œ„์˜ ์‹์—์„œ๋„ $\displaystyle \lim_{y\rightarrow 0} M_Z(yt)โ€™ = M_Z(0)โ€™ = 0 = \mu$์ด๋ฏ€๋กœ, ๋‹ค์‹œ โ€œ๋กœํ”ผํƒˆ ์ •๋ฆฌโ€๋ฅผ ์ ์šฉํ•˜๋ฉด,

\[\begin{aligned} \frac{t}{2} \cdot \lim_{y \rightarrow 0} \frac{M_z(yt)'}{y} &= \frac{t}{2} \cdot \lim_{y \rightarrow 0} \frac{t M_z (yt) ''}{1} \\ &= \frac{t^2}{2} \cdot \lim_{y \rightarrow 0} M_z (yt) '' \\ &= \frac{t^2}{2} \quad \left(\because \; \lim_{y \rightarrow 0} M_z (yt) '' = 1 = \sigma^2\right) \end{aligned}\]

๋”ฐ๋ผ์„œ,

\[\lim_{n\rightarrow \infty} \log M_W(t) = \frac{t^2}{2}\]

์œ„์˜ ์‹์—์„œ $\log$๋ฅผ ์‹œ์ผœ์„œ $\bar{X}$์˜ ์ •๊ทœํ™”์ธ $W$์˜ mgf $M_W(t)$๋ฅผ ์–ป์œผ๋ฉด ์•„๋ž˜์™€ ๊ฐ™๋‹ค.

\[\lim_{n\rightarrow \infty} M_W(t) = e^{t^2/2}\]

์ด๋•Œ, ํ‘œ์ค€์ •๊ทœ๋ถ„ํฌ $N(0, 1)$์˜ mgf๊ฐ€ $e^{t^2/2}$์ด๊ณ , ๋‘ ๋ถ„ํฌ์˜ mgf๊ฐ€ ๊ฐ™์œผ๋ฏ€๋กœ, โ€œUniqueness of mgfโ€์— ์˜ํ•ด ์•„๋ž˜์˜ ๋ช…์ œ๊ฐ€ ์„ฑ๋ฆฝํ•œ๋‹ค.

โ€œ$n$์ด ์ถฉ๋ถ„ํžˆ ์ปค์ง€๋ฉด, $\bar{X}$์˜ ์ •๊ทœํ™”์ธ $\dfrac{\bar{X} - \mu}{\sigma/\sqrt{n}}$๋Š” ํ‘œ์ค€์ •๊ทœ๋ถ„ํฌ $N(0, 1)$์„ ๋”ฐ๋ฅธ๋‹ค!โ€

$\blacksquare$


Sampling Distribution of the difference btw two mean

์ด๋ฒˆ์—๋Š” ๋‘ ๊ฐœ์˜ ์„œ๋กœ population์—์„œ ๋ฝ‘์€ ๋‘ independent sample์„ ์ƒ๊ฐํ•ด๋ณด์ž!

Let $X_1, \dots, X_{n_1}$, and $Y_1, \dots, Y_{n_2}$ be two independent random samples with $E[X_1] = \mu_1$, $\text{Var}(X_1) = \sigma_1^2$, and $E[X_2] = \mu_2$, $\text{Var}(Y_2) = \sigma_2^2$.

์šฐ๋ฆฌ๋Š” โ€œ๋‘ ์ƒ˜ํ”Œ ํ‰๊ท ์˜ ์ฐจโ€ $\mu_1 - \mu_2$์— ๋Œ€ํ•œ ๋ถ„ํฌ๋ฅผ ๋ชจ๋ธ๋งํ•˜๊ณ ์ž ํ•œ๋‹ค. ์ด๋•Œ, $\overline{X} - \overline{Y}$๋ฅผ ์‚ฌ์šฉํ•˜๋ฉด โ€œ๋‘ ์ƒ˜ํ”Œ ํ‰๊ท ์˜ ์ฐจโ€์— ๋Œ€ํ•ด ์ถ”๋ก ํ•  ์ˆ˜ ์žˆ๋‹ค!!

By CLT,

\[\begin{aligned} \frac{\overline{X} - \mu_1}{\sigma_1/\sqrt{n_1}} \sim N(0, 1) \quad & \iff \quad \overline{X} \sim N\left(\mu_1, \sigma_1^2/n_1\right) \\ \frac{\overline{Y} - \mu_2}{\sigma_2/\sqrt{n_2}} \sim N(0, 2) \quad & \iff \quad \overline{Y} \sim N\left(\mu_2, \sigma_2^2/n_2\right) \end{aligned}\]

๋”ฐ๋ผ์„œ, $\overline{X} - \overline{Y}$์— ๋Œ€ํ•œ ๋ถ„ํฌ๋Š” independentํ•œ ๋‘ normal distribution์— ๋Œ€ํ•œ ๋ง์…ˆ์œผ๋กœ ์‰ฝ๊ฒŒ ์œ ๋„ํ•  ์ˆ˜ ์žˆ๋‹ค!

\[\overline{X} - \overline{Y} \sim N\left( \mu_1 - \mu_2, \; \frac{\sigma_1^2}{n_1} + \frac{\sigma_2^2}{n_2} \right)\]

์œ„์˜ ์‚ฌ์‹ค์„ ์ด์šฉํ•ด์„œ, โ€œ๋‘ ์ƒ˜ํ”Œ ํ‰๊ท ์˜ ์ฐจโ€์— ๋Œ€ํ•œ ์ถ”๋ก ๋„ ์‰ฝ๊ฒŒ ์ˆ˜ํ–‰ํ•  ์ˆ˜ ์žˆ๋‹ค ๐Ÿ˜‰


๋งบ์Œ๋ง

์ด๋ฒˆ ํฌ์ŠคํŠธ์—์„œ๋Š” ํ‘œ๋ณธํ‰๊ท  $\bar{X}$์— ๋Œ€ํ•œ ๋ถ„ํฌ์ธ โ€œSampling Distribution of Meanโ€์„ ๋ณด์•˜๋‹ค. ๋˜, ํ‘œ๋ณธํ‰๊ท  $\bar{X}$์˜ ๋ถ„ํฌ๋ฅผ ํŒŒ์•…ํ•˜๊ณ , ํ™œ์šฉํ•˜๋Š”๋ฐ ํ•„์š”ํ•œ <WLLN>๊ณผ <CLT>๋ฅผ ์‚ดํŽด๋ณด์•˜๋‹ค.

์ด์–ด์ง€๋Š” ํฌ์ŠคํŠธ์—์„œ๋Š” โ€œํ‰๊ท โ€๊ณผ ํ•จ๊ป˜, ํ™•๋ฅ  ๋ถ„ํฌ์˜ ํŠน์„ฑ์„ ๊ฒฐ์ •ํ•˜๋Š” parameter์ธ โ€œ๋ถ„์‚ฐ(Variance)โ€์ด Random Sample์—์„œ ์–ด๋–ป๊ฒŒ ์œ ๋„๋˜๋Š”์ง€ ์‚ดํŽด๋ณผ ์˜ˆ์ •์ด๋‹ค.

๐Ÿ‘‰ Sampling Distribution of Variance


references