intro_2_ggplot2.Rmd

---
title: "tidyverse: ggplot2"
author: 郭耀仁
output: 
  revealjs::revealjs_presentation:
    theme: black
    highlight: zenburn
    center: true
---

# ggplot2 套件

## 超級受歡迎

- 包含於 tidyverse 之中
- 視覺化的文法
- 將變數指定在 `aes()` 之中

```{r eval=FALSE}
ggplot(df, aes()) +
  geom_xxx()
```

## 基本的統計圖形

- 散佈圖：`geom_point()`
- 線圖：`geom_line()`
- 直方圖：`geom_histogram()`
- 盒鬚圖：`geom_boxplot()`
- 長條圖：`geom_bar()`

## 繪圖前先（安裝）載入這些套件

```{r eval=FALSE}
install.packages(c("tidyverse", "gapminder"))
```

```{r message=FALSE}
library(tidyverse)
library(gapminder)
```

## 散佈圖

```{r}
gapminder_2007 <- gapminder %>%
  filter(year == 2007)
scatter <- ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp)) +
  geom_point() +
  theme_minimal()
```

---

```{r}
scatter
```

## 線圖

```{r}
north_asia <- gapminder %>%
  filter(country %in% c("China", "Japan", "Taiwan", "Korea, Rep."))
line_plot <- ggplot(north_asia, aes(x = year, y = gdpPercap, colour = country)) +
  geom_line() +
  theme_minimal()
```

---

```{r}
line_plot
```

## 直方圖

```{r}
histogram <- ggplot(gapminder_2007, aes(x = gdpPercap)) +
  geom_histogram(bins = 20) +
  theme_minimal()
```

---

```{r}
histogram
```

## 盒鬚圖

```{r}
box_plot <- ggplot(gapminder_2007, aes(x = continent, y = gdpPercap)) +
  geom_boxplot() +
  theme_minimal()
```

---

```{r}
box_plot
```

## 長條圖

```{r}
gdpPercap_2007_na <- gapminder %>%
  filter(year == 2007 & country %in% c("China", "Japan", "Taiwan", "Korea, Rep."))

bar_plot <- ggplot(gdpPercap_2007_na, aes(x = country, y = gdpPercap)) +
  geom_bar(stat = "identity") +
  theme_minimal()
```

---

```{r}
bar_plot
```

# 隨堂練習

## 練習資料集

- [點此下載](https://storage.googleapis.com/learn-r-the-easy-way.appspot.com/hh.xlsx)

```{r eval=FALSE}
install.packages("readxl")
library(readxl)
hh <- read_excel("YOUR_EXCEL_PATH")
```

```{r echo=FALSE}
library(readxl)
hh <- read_excel("~/Downloads/hh.xlsx")
```

## 欄位定義

- district: 行政區
- hh: 戶數
- disposable_income_in_thousands: 可支配所得（新台幣千元）
- year: 年份

## 大安區、中正區與信義區的平均每戶可支配所得線圖

```{r echo=FALSE}
line_df <- hh %>%
  filter(district %in% c("大安區", "中正區", "信義區")) %>%
  mutate(avg_income = disposable_income_in_thousands * 1000 / hh)

ggplot(line_df, aes(x = year, y = avg_income, colour = district)) +
  geom_line() +
  theme_minimal() +
  theme(text = element_text(family = 'Heiti TC Light'))
```

## 所有行政區的每戶可支配所得盒鬚圖

```{r echo=FALSE}
box_df <- hh %>%
  mutate(avg_income = disposable_income_in_thousands * 1000 / hh)

ggplot(box_df, aes(x = district, y = avg_income)) +
  geom_boxplot() +
  theme_minimal() +
  theme(text = element_text(family = "Heiti TC Light"))
```

## 平均每戶可支配所得的最新排名

```{r echo=FALSE}
bar_df <- hh %>%
  filter(year == 2016) %>%
  mutate(avg_income = disposable_income_in_thousands * 1000 / hh) %>%
  arrange(avg_income)

district_levels <- as.character(bar_df$district)
bar_df$district <- factor(bar_df$district, levels = district_levels)

ggplot(bar_df, aes(x = district, y = avg_income)) +
  geom_bar(stat = "identity") +
  theme_minimal() +
  theme(text = element_text(family = "Heiti TC Light")) +
  coord_flip()
```

# 延伸閱讀

## 官方文件

http://ggplot2.tidyverse.org/reference/index.html