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

5 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


Definition. F-distribution

If V1โˆผฯ‡2(n1) and V2โˆผฯ‡2(n2) are independent,

then F:=V1/n1V2/n2 is called <Snedecorโ€™s F-distribution>1 with degrees of freedom n1 and n2, and denoted as

FโˆผF(n1,n2)

ps) ์ผ๋ฐ˜์ ์œผ๋กœ, F(n1,n2)โ‰ F(n2,n1)์ด๋‹ค. F-distribution์€ non-symmetric์ด๋ผ๋Š” ๋ง.

Image from Wikipedia


Remark.

1. The order of n1 and n2 is very important.

In fact we have F(n1,n2)=D1F(n2,n1).

2. Let fฮฑ(n1,n2) be the number x such that ฮฑ=P(F(n1,n2)โ‰ฅx).

Here, we have f1โˆ’ฮฑ(n1,n2)=1fฮฑ(n2,n1)

Quick Proof.

1โˆ’ฮฑ=P(F(n1,n2)โ‰ฅf1โˆ’ฮฑ(n1,n2))=P(1f1โˆ’ฮฑ(n1,n2)โ‰ฅ1F(n1,n2))=P(1f1โˆ’ฮฑ(n1,n2)โ‰ฅF(n2,n1))=1โˆ’P(F(n2,n1)>1f1โˆ’ฮฑ(n1,n2))

๋”ฐ๋ผ์„œ,

ฮฑ=P(F(n2,n1)>1f1โˆ’ฮฑ(n1,n2))=P(F(n2,n1)>fฮฑ(n2,n1))

๋”ฐ๋ผ์„œ,

fฮฑ(n1,n2)=1f1โˆ’ฮฑ(n2,n1)

โ—ผ


Theorem.

Supp. we have two independent random samples X1,โ€ฆ,Xn1 from N(ฮผ1,ฯƒ12) and Y1,โ€ฆ,Yn2 from N(ฮผ2,ฯƒ22).

Let S12=โˆ‘i=1n1(Xiโˆ’Xยฏ)2n1โˆ’1 and S22=โˆ‘i=1n2(Yiโˆ’Yยฏ)2n2โˆ’1.

Note that (n1โˆ’1)S12/ฯƒ12โˆผฯ‡2(n1โˆ’1) and (n2โˆ’1)S22/ฯƒ22โˆผฯ‡2(n2โˆ’1).

Then,

F:=S12/ฯƒ12S22/ฯƒ22โˆผF(n1โˆ’1,n2โˆ’1)

Proof.

F:=S12/ฯƒ12S22/ฯƒ22=(n1โˆ’1)S12ฯƒ12/(n1โˆ’1)(n2โˆ’1)S22ฯƒ22/(n2โˆ’1)

์ด๋•Œ, (n1โˆ’1)S12ฯƒ12โˆผฯ‡2(n1โˆ’1)์ด๋ฏ€๋กœ <F-distribution>์˜ ์ •์˜์— ๋”ฐ๋ผ

F:=S12/ฯƒ12S22/ฯƒ22=V1/(n1โˆ’1)V2/(n2โˆ’1)โˆผF(n1โˆ’1,n2โˆ’1)

ExamplesPermalink

n1=21, n2=31

Claim: ฯƒ12/ฯƒ22=2 but, for sample variances, S12/S22=4>2.

P(S12/S22โ‰ฅ4whenฯƒ12/ฯƒ22=2)=P(S12/ฯƒ12S22/ฯƒ22โ‰ฅ4โ‹…12=2)=P(F(20,30)โ‰ฅ2)

Here, f0.05(20,30)=1.93 and f0.01(20,30)=2.55.

The the value of 2 is btw 1.93 and 2.55.

Therefore,

P(F(20,30)โ‰ฅ2)โˆˆ[0.01,0.05]

์ด๊ฒƒ์˜ ์˜๋ฏธ๋Š” sample variance์˜ ๋น„์œจ์ด 4๊ฐ€ ๋˜๋Š” ํ™•๋ฅ ์€ ์ง€๊ทนํžˆ ๋‚ฎ๋‹ค๋Š” ๊ฒƒ์ด๋‹ค. ๊ทธ๋Ÿฐ๋ฐ ์ด๊ฒƒ์ด ์‹ค์ œ๋กœ ๊ด€์ธก๋˜์—ˆ์œผ๋ฏ€๋กœ, ์šฐ๋ฆฌ์˜ ๊ฐ€์ •์ธ H0:ฯƒ12/ฯƒ22=2๋ฅผ ๊ธฐ๊ฐํ•˜๊ณ , ๋‘˜์˜ population variance์˜ ๋น„์œจ์ด ๋” ์ปค์ ธ์•ผ ํ•œ๋‹ค๋Š” ๋Œ€๋ฆฝ ๊ฐ€์„ค H1:ฯƒ12/ฯƒ22>2๋ฅผ ์ฑ„ํƒํ•ด์•ผ ํ•œ๋‹ค. โ—ผ


์ง€๊ธˆ๊นŒ์ง€ ์šฐ๋ฆฌ๋Š” population distribution์˜ parameter์ธ โ€œํ‰๊ท โ€๊ณผ โ€œ๋ถ„์‚ฐโ€์— ๋Œ€ํ•ด ์ถ”์ •ํ–ˆ๋‹ค. ์ด์–ด์ง€๋Š” ํฌ์ŠคํŠธ์—์„œ๋Š” sample๋กœ๋ถ€ํ„ฐ ์–ป๋Š” ๋ถ„ํฌ์ธ <EDF; Empirical Distribution Function>์œผ๋กœ๋ถ€ํ„ฐ population distribution์„ ์ถ”์ •ํ•ด๋ณธ๋‹ค. ์ด ๊ณผ์ •์—์„œ ์“ฐ๋Š” ๊ฒƒ์ด ๋ฐ”๋กœ <Quantile; ๋ถ„์œ„์ˆ˜>์ด๋‹ค!

๐Ÿ‘‰ EDF and Quantile


  1. โ€œ[์„ธ๋„ค๋ฐ์ปค] F-๋ถ„ํฌโ€๋ผ๊ณ  ์ฝ๋Š” ๊ฒƒ ๊ฐ™๋‹ค. โ†ฉ