Possion Distribution์—์„œ ๐œ†๋ฅผ ๊ตฌํ•˜๊ธฐ

๋ฌธ์ œ

  • ISP(Internet Service Provider์— t ์‹œ์ ์— ๋„๋‹ฌํ•˜๋Š” Ping์˜ ์ˆ˜๋ฅผ $X_t$๋ผ๊ณ  ํ•˜
  • ์ด ๋•Œ $X_t$๋Š” Random Variable์ด๊ณ  ๋งค 10์ดˆ๋งˆ๋‹ค Ping์˜ ์—†๋Š” ํ™•๋ฅ ์„ 0.001์ด๋ผ๊ณ  ํ•  ๋•Œ, 30์ดˆ ๋‚ด์— Ping์ด 10๋ฒˆ์ด์ƒ ์˜ฌ ํ™•๋ฅ ์€ ์–ผ๋งˆ์ธ๊ฐ€?

ํ•ด๊ฒฐ

  • $X_{t} \sim \text{Poisson}(\lambda t)$ ํ˜•ํƒœ๋กœ ์‹์„ ์ •๋ฆฌํ•  ์ˆ˜ ์žˆ๋‹ค.
  • $0.001 = P(X_{10}=0) = \frac{\exp(-10\lambda)(10\lambda)^{0}}{0!} = \exp(-10\lambda) $ ์ด๋ฏ€๋กœ $\lambda = \frac{-\log(0.001)}{10} $์ด๋‹ค.
  • ์ด๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ 30์ดˆ ๋‚ด์— 0๋ฒˆ์ผ ํ™•๋ฅ ์„ ๊ตฌํ•˜๋ฉด ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค. ์ด๋•Œ Poisson Distribution์˜ PDF์‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™์œผ๋‹ˆ ํ•ด๋‹น ์‹์— ๊ฐ’์„ ๋„ฃ์–ด์„œ $\lambda$๋ฅผ ๊ตฌํ•ด์ค€๋‹ค.

$$P(X=x)=e^{-\lambda }\cdot \frac{\lambda ^x}{x!}$$

$$P(X_{30}=0) = \frac{\exp(3\log(0.001))(-3\log(0.001))^{0}}{0!} ย = (0.001)^{3}$$

  • ์—ฌ๊ธฐ์„œ ์œ ์ถ”ํ•  ์ˆ˜ ์žˆ๋Š” ํŠน์„ฑ์€ ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค. ${30 \over ย 10}=3$์ด๋ผ์„œ ์ง€์ˆ˜๋กœ 3์ด ์œ„ ์‹์— ๋ถ™์€ ๊ฒƒ์„ ์•Œ ์ˆ˜ ์žˆ๋‹ค.

$$P(X_{t_2}=0)=P(X_{t_1}=0)^{\frac{t_2}{t_1}}$$

  • ์ด๋ฅผ ๋ฐ”ํƒ•์œผ๋กœ 30์ดˆ ๋‚ด์— Ping์ด 10๋ฒˆ ์ด์ƒ ์˜ฌ ํ™•๋ฅ ์„ ๊ตฌํ•  ์ˆ˜ ์žˆ๋‹ค. ์ด๋•Œ $\Phi$๋ฅผ ํ‘œ์ค€์ •๊ทœ๋ถ„ํฌ์˜ CDF(Cumulative Distribution Function)๋ผ๊ณ  ํ•œ๋‹ค๋ฉด ์ „์ฒด ์‹์€ ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค. Poission Distribution์˜ Mean๊ณผ Variance๋Š” $\lambda$๋กœ ๋™์ผํ•˜๋‹ค.

$$P(X_{30} \geq 10) = P\left(\frac{X_{30}-E[X_{30}]}{\sqrt{Var[X_{30}]}} \geq \frac{10-E[X_{30}]}{Var[X_{30}]} \right) \\ = P\left(\frac{X_{30}-E[X_{30}]}{\sqrt{Var[X_{30}]}} \geq \frac{10-30\lambda}{\sqrt{30\lambda}} \right) \\ \approx 1-\Phi\left(\frac{10-30\lambda}{\sqrt{30\lambda}}\right) \\ \approx 1-\Phi\left(-2.36\right) \\ \approx 1-0.009=0.991$$

References

Read more

DataFrame์€ Pandera๋กœ, ๋ชจ๋ธ์€ Pydantic์œผ๋กœ ๋ฐ์ดํ„ฐ๋ฅผ ๊ฒ€์ฆํ•œ๋‹ค.

DataFrame์€ Pandera๋กœ, ๋ชจ๋ธ์€ Pydantic์œผ๋กœ ๋ฐ์ดํ„ฐ๋ฅผ ๊ฒ€์ฆํ•œ๋‹ค.

Pandera: ๋ฐ์ดํ„ฐํ”„๋ ˆ์ž„ ๊ฒ€์ฆ์— ์ตœ์ ํ™”๋œ ๋„๊ตฌ ์ฃผ์š” ์žฅ์  * Pandas์™€ ํ†ตํ•ฉ: Pandas ๋ฐ์ดํ„ฐํ”„๋ ˆ์ž„์— ๋Œ€ํ•ด ์Šคํ‚ค๋งˆ ๊ธฐ๋ฐ˜ ๊ฒ€์ฆ์„ ์ˆ˜ํ–‰ํ•ฉ๋‹ˆ๋‹ค. * ์œ ์—ฐํ•œ ๊ฒ€์ฆ ์กฐ๊ฑด: ์—ด ๋ฐ์ดํ„ฐ ํƒ€์ž…, ๊ฐ’ ๋ฒ”์œ„, Null ์—ฌ๋ถ€ ๋“ฑ ๋‹ค์–‘ํ•œ ๊ฒ€์ฆ ์กฐ๊ฑด์„ ์ •์˜ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. * ๋ช…ํ™•ํ•œ ์˜ค๋ฅ˜ ๋ฉ”์‹œ์ง€: ์Šคํ‚ค๋งˆ ๋ถˆ์ผ์น˜์— ๋Œ€ํ•œ ๋ช…ํ™•ํ•œ ์˜ค๋ฅ˜ ๋ฉ”์‹œ์ง€๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ๋‹จ์  * ๋Œ€์šฉ๋Ÿ‰ ๋ฐ์ดํ„ฐ ๊ฒ€์ฆ์—์„œ๋Š” ์†๋„๊ฐ€ ๋Š๋ฆด ์ˆ˜

Docker ๋ชจ๋‹ˆํ„ฐ๋งํ•˜๋ฉด์„œ ์ฃฝ์œผ๋ฉด ์žฌ์‹œ์ž‘์‹œํ‚ค๋Š” ์Šคํฌ๋ฆฝํŠธ

Docker ๋ชจ๋‹ˆํ„ฐ๋งํ•˜๋ฉด์„œ ์ฃฝ์œผ๋ฉด ์žฌ์‹œ์ž‘์‹œํ‚ค๋Š” ์Šคํฌ๋ฆฝํŠธ

Docker ์žฌ์‹œ์ž‘ ์Šคํฌ๋ฆฝํŠธ #!/bin/bash CONTAINERS=("container1_name" "container2_name") # ๊ฐ์‹œํ•  ์ปจํ…Œ์ด๋„ˆ ์ด๋ฆ„ ์ž…๋ ฅ CHECK_INTERVAL=600 # ์ฒดํฌ ์ฃผ๊ธฐ (์ดˆ ๋‹จ์œ„) while true; do for CONTAINER_NAME in "${CONTAINERS[@]}"; do # ์ปจํ…Œ์ด๋„ˆ ์ƒํƒœ ํ™•์ธ CONTAINER_STATUS=$(docker inspect -f '{{.State.Status}}' "$CONTAINER_

Tobit Regression์€ Censored Data์— ์ ํ•ฉํ•œ Regression์ด๋‹ค.

Tobit Regression์€ Censored Data์— ์ ํ•ฉํ•œ Regression์ด๋‹ค.

Tobit Regression * Tobit ํšŒ๊ท€(Tobit Regression)๋Š” ์ข…์† ๋ณ€์ˆ˜๊ฐ€ ํŠน์ • ๊ฐ’์—์„œ ์ ˆ๋‹จ(Censored)๋œ ์ƒํ™ฉ์—์„œ ๋ฐ์ดํ„ฐ๋ฅผ ๋ถ„์„ํ•˜๊ธฐ ์œ„ํ•ด ์‚ฌ์šฉ๋˜๋Š” ํ†ต๊ณ„ ๊ธฐ๋ฒ•์ž…๋‹ˆ๋‹ค. * James Tobin์ด ์ฒ˜์Œ ์ œ์•ˆํ•œ ์ด ๋ชจ๋ธ์€ ๊ฒฝ์ œํ•™๊ณผ ์‚ฌํšŒ๊ณผํ•™ ๋ถ„์•ผ์—์„œ ์ž์ฃผ ์‚ฌ์šฉ๋˜๋ฉฐ, ์ผ๋ฐ˜ ์„ ํ˜• ํšŒ๊ท€๋กœ๋Š” ์„ค๋ช…ํ•  ์ˆ˜ ์—†๋Š” ์ƒํ™ฉ์—์„œ ํšจ๊ณผ์ ์œผ๋กœ ์ ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. Tobit Regression ์ˆ˜์‹ 1. ๊ด€์ธก๋œ ์ข…์† ๋ณ€์ˆ˜