###### tags: `건국대` `챗GPT 디지털 글쓰기` `한국R사용자회`

# 8/12 (건국대)

# https://bit.ly/45rklmU

- 소프트웨어 카펜트리
- https://statkclee.github.io/2023-08-05-konkuk/
- 챗GPT 유닉스 쉘
- https://ebook-product.kyobobook.co.kr/dig/epd/ebook/E000005358063

- 데이터
- https://swcarpentry.github.io/shell-novice/
- download shell-lesson-data.zip

---
title: "재미날것 같은 통계학"
format: html
editor: visual
editor_options:
chunk_output_type: console
---

# 통계학

## 추론통계

## 요약통계

### 범주형 변수

library(tidyverse)
library(palmerpenguins)
library(janitor)

penguins |> 
  drop_na() |> 
  count(species, sex, sort = TRUE) |> 
  pivot_wider(names_from = sex, values_from = n) |> 
  mutate(합계 = female + male)  |> 
  janitor::adorn_totals(where = "row", name = "합계") |> 
  gt::gt()

### 연속형 변수

weights_tbl <- penguins |> 
  drop_na() |> 
  group_by(species) |> 
  summarise(평균체중 = mean(body_mass_g),
            SD체중  = sd(body_mass_g),
            최소체중 = min(body_mass_g),
            최대체중 = max(body_mass_g))

weights <- weights_tbl |> 
  pull(평균체중)

weights

남극에 서식중인 펭귄 3종의 평균 체중은 다음과 같습니다.

- 아델리: `r scales::comma(weights[1])`
- 턱끈: `r weights[2]`
- 젠투: `r weights[3]`

# 시각화

## 상자그림

penguins |> 
  ggplot(aes(x = species, y = body_mass_g)) +
    geom_boxplot() +
    geom_jitter(aes(color = species))

## 빈도수

penguins |>
  drop_na() |> 
  count(species, sex) |> 
  ggplot(aes(x = fct_reorder(species, -n), y = n, fill = sex)) +
    geom_col(width=0.5) +
    labs(x = "펭귄 종", y = "펭귄 빈도수",
         fill = "암수",
         title = "남극 서식 펭귄종 통계 빈도 분석") +
    theme_minimal(base_family = "AppleGothic")

# 표본추출

## SRS

srs_penguins <- penguins |> 
  drop_na() |> 
  slice_sample(n = 100)

## 집락추출

cluster_penguins <- penguins |> 
  drop_na() |> 
  mutate(cluster = cut(bill_length_mm, breaks = 10)) 

# 1. 단계 집락추출
clusters <- unique(cluster_penguins$cluster)
clusters_sample <- sample(clusters, size = 3)

# 2. 펭귄 추출
cluster_penguins |> 
  filter(cluster %in% clusters_sample) |> 
  slice_sample(n = 10)

## 표본평균

penguins |> 
  drop_na() |> 
  summarise(표본평균체중 = mean(body_mass_g ))

trials <- rep(100, 1000)

get_srs <- function(n_sample = 100) {
  srs_penguins <- penguins |> 
    slice_sample(n = n_sample)
  
  return(srs_penguins)
}

get_srs(100)

get_mean <- function(dataframe) {
  dataframe |>
    summarise(표본평균체중 = mean(body_mass_g, na.rm = TRUE )) |> 
    pull(표본평균체중)
}


trials |> 
  enframe() |> 
  mutate(sample = map(value, get_srs)) |> 
  mutate(sample_mean = map_dbl(sample, get_mean)) |> 
  summarise(mean(sample_mean))

penguins |> 
  drop_na() |> 
  summarise(표본평균체중 = mean(body_mass_g ))