Japan <- gapminder %>%
  dplyr::filter(country == "Japan") %>%
  dplyr::select(year, lifeExp) # year と lifeExp だけを抜き出す

Japan # 抜き出したデータの中身を表示して確認

ggplot(Japan, aes(x = year, y = lifeExp)) +
  geom_point() +
  geom_line()

jpn.chi <- gapminder %>%
  dplyr::filter(country == "China" | country == "Japan") %>%
  dplyr::select(year, country, lifeExp)　# year、country、lifeExp だけを抜き出す

jpn.chi

DT::datatable(jpn.chi)

ggplot(jpn.chi, aes(x = year, y = lifeExp, color = country)) +
  geom_point() +
  geom_line() +
  ggtitle("Life Expectancy: Japan and China 1952-2007")

summary(hr$age)

party.median <- hr %>%                                # party.medianとして保存
  dplyr::filter(party == "LDP" | party == "DPJ") %>%  # 自民党と民主党だけを残す
  dplyr::filter(wl == 1) %>% 　　　　　　　　　　     　　# 当選者 (wl = 1) だけを残す
  drop_na(age) %>%                                    # ageの5つの欠損値をドロップ
  group_by(year, party) %>%                           # year毎、party毎に計算する  
  summarise(age.median = median(age))   # ageの中央値の平均を age.maeian として保存

DT::datatable(party.median)

ggplot(party.median, aes(x = year, y = age.median, 
                         color = party, linetype = party, shape = party)) +
  geom_point() +
  geom_line() +
  lims(y = c(40, 60)) +    　　# y 軸の範囲を指定
  ggtitle("政党別当選者年齢の中央値 (1996-2014)") +
  scale_color_discrete(name ="政党",
                       labels = c("民主党", "自民党")) +
  scale_linetype_discrete(name ="政党",
                          labels = c("民主党", "自民党")) + 
  scale_shape_discrete(name ="政党",
                          labels = c("民主党", "自民党")) +　　
  ggtitle("当選者年齢の推移 (自民党と民主党：1996年〜2014年)") +
  labs(x = "総選挙年", y = "年齢の中央値")　

US <- US %>%                                                    　# US として保存
  dplyr::filter(party == "Republican" | party == "Democrat") %>%  # Rep と Dem だけを残す
  group_by(congress, party) %>%                                  # congress毎、party毎に計算する  
  summarise(econ.median = median(dwnom1))         # dwnom1 の中央値の平均を econ.maeian として保存

DT::datatable(US)

ggplot(US, aes(x = congress, y = econ.median, color = party)) +
  geom_point() +
  geom_line() +
  theme_bw()　+               # 背景色を白にしたい場合 
  ggtitle("Political Polarization: ECON Dimention(US Congress:1947-2012)") +
  labs(x = "Congress",　　　　# x 軸のラベルを指定
       y = "DW-NOMINATE score (economic dimention)")　# y 軸のラベルを指定

library("tidyverse")
library("jpndistrict")
library("sf")
library("magrittr")
library("ggimage")
library("DT")

download.file(url = "https://population.un.org/wpp/Download/Files/1_Indicators%20(Standard)/CSV_FILES/WPP2019_TotalPopulationBySex.csv",
              destfile = "data/world_pop.csv")

df_world_pop <- read_csv("data/world_pop.csv") %>% 
    filter(Time == 2016) %>%       
    rename(country = Location) %>%             # 変数Locationをcountryに変更
    filter(!str_detect(country, "China, "), LocID != 1105,
           LocID != 850, LocID != 1111) %>% 
    mutate(
        country = str_remove_all(country, c(" of America")), # of America の部分を削除 
        country = str_replace_all(country, c("Russian Federation" = "Russia")) # 名称を簡略化  
        ) %>% 
    select(Time, LocID, country, PopTotal) # 4 つの変数だけを選ぶ  

DT::datatable(df_world_pop)

df_world_sf <- rnaturalearth::ne_countries(returnclass = "sf") %>% 
    rename(country = sovereignt) %>% 
    select(country, geometry) %>% 
    filter(country != "Antarctica") %>% 
    mutate(
        country = str_remove_all(country, c(" of America"))
        )

head(df_world_sf)

df_world <- df_world_pop %>% 
  merge(df_world_sf, by = "country", all = T) %>% 
  distinct(LocID, country,.keep_all = TRUE) %>% 
  rownames_to_column() %>% 
  st_as_sf()

names(df_world)

map_pop_world <- df_world %>% 
  ggplot() + 
  geom_sf(aes(fill = PopTotal)) +
  scale_fill_distiller(name = "人口",
                       palette = "Reds", direction = 1, 
                       na.value = "grey", guide = "legend") +
  theme_map(base_family = "HiraginoSans-W3") +
  theme(legend.position = c(0.1, 0.3),
        legend.title = element_text(size = 45), 
        legend.text = element_text(size = 40),
        legend.key.size = unit(5, "cm"),
        legend.key.width = unit(4.5,"cm")) +
  coord_sf(datum = NA) 

map_pop_world

ggsave("fig/map_pop_world.png", map_pop_world, width = 45, height = 45)

library(jpndistrict)

df_sf_ja <- 1:47 %>%
  map(~ jpn_pref(pref_code = ., district = FALSE)) %>% 
  reduce(rbind) %>% 
  st_simplify(dTolerance = 0.01)

df_sf_ja_omit47 <- df_sf_ja %>% 
  filter(pref_code != "47")
df_sf_ja_okinawa <- df_sf_ja %>%
  filter(pref_code == "47")
df_sf_ja_okinawa$geometry %<>%
  add(c(5.6, 17.5))
df_sf_ja_okinawa %<>% st_set_crs(value = 4326)

df_sf_ja<- df_sf_ja %>% 
  filter(pref_code != "47") %>% 
  bind_rows(df_sf_ja_okinawa)

names(df_sf_ja)

df_jpn_pop <- read_csv("data/jpn_pop.csv", 
                       locale = locale(encoding = "cp932"))

df_jap <- df_jpn_pop %>% 
    left_join(df_sf_ja, by = "prefecture") %>%   
    st_as_sf()

head(df_jap)

map_pop_jpn_1 <- df_jap %>% 
    ggplot() +
    geom_sf(aes(fill = total)) +                                # total を指定
    geom_segment(aes(x = round(st_bbox(df_sf_ja_omit47)[1], 0),
                     xend = 132.5, y = 40, yend = 40)) + 
    geom_segment(aes(x = 132.5, xend = 138, 
                     y = 40, yend = 42)) +
    geom_segment(aes(x = 138, xend = 138,
                     y = 42, yend = round(st_bbox(df_sf_ja_omit47)[4],  0))) +
    scale_fill_distiller(name = "人口",
                         palette = "Greens", direction = 1) +   # 任意の色を指定
    theme_map(base_family = "HiraginoSans-W3") +
    theme(legend.position = c(0.8, 0.5),
          legend.title = element_text(size = 55), 
          legend.text = element_text(size = 50),
          legend.key.size = unit(5, "cm"),
          legend.key.width = unit(4.5,"cm")) +
    coord_sf(datum = NA) 

map_pop_jpn_1

ggsave("fig/map_pop_jpn_1.png", map_pop_jpn_1, width = 45, height = 45)

map_pop_jpn_2 <- df_jap %>% 
    ggplot() +
    geom_sf(aes(fill = diff)) +                                  # diff を指定
    geom_segment(aes(x = round(st_bbox(df_sf_ja_omit47)[1], 0),
                     xend = 132.5, y = 40, yend = 40)) + 
    geom_segment(aes(x = 132.5, xend = 138, 
                     y = 40, yend = 42)) +
    geom_segment(aes(x = 138, xend = 138,
                     y = 42, yend = round(st_bbox(df_sf_ja_omit47)[4],  0))) +
    scale_fill_distiller(name = "人口増減(%)",
                         palette = "RdBu", direction = 1) +　　　# 色を赤と青に指定
    theme_map(base_family = "HiraginoSans-W3") +
    theme(legend.position = c(0.8, 0.5),
          legend.title = element_text(size = 55), 
          legend.text = element_text(size = 50),
          legend.key.size = unit(5, "cm"),
          legend.key.width = unit(4.5,"cm")) +
    coord_sf(datum = NA) 

map_pop_jpn_2

ggsave("fig/map_pop_jpn_2.png", map_pop_jpn_2, width = 45, height = 45)

library(jpndistrict)

head(df_sf_ja, 23)

df_tokyo_sf <- jpn_pref(13, district = TRUE) %>% 
  filter(str_detect(city, "区"))

head(df_tokyo_sf, 23)

df_tokyo_pop <- read_csv("data/tokyo_pop.csv")

head(df_tokyo_sf)

head(df_tokyo_pop)

df_tokyo_sf <- df_tokyo_sf %>% 
    left_join(df_tokyo_pop, by = "city") %>% 
    st_as_sf()

df_tokyo_sf %>% 
    head() %>% 
    rmarkdown::paged_table()

map_pop_tokyo <- df_tokyo_sf %>% 
    ggplot() +
    geom_sf(aes(fill = population)) +
    scale_fill_distiller(name = "人口",
                         palette = "Greens", direction = 1) +
    theme_map(base_family = "HiraginoSans-W3") +
    theme(legend.position = c(.1, -.1),
          legend.direction = "horizontal",
          legend.title = element_text(size = 15), 
          legend.text = element_text(size = 15),
          legend.key.size = unit(1, "cm"),
          legend.key.width = unit(3,"cm")) +
    coord_sf(datum = NA) 

map_pop_tokyo

ggsave("fig/map_pop_tokyo.png", map_pop_tokyo, width = 10, height = 10)

map_pop_tokyo_text <- df_tokyo_sf %>% 
   mutate(
        text_x = map_dbl(geometry, ~st_centroid(.x)[[1]]),
        text_y = map_dbl(geometry, ~st_centroid(.x)[[2]])
        ) %>% 
  ggplot() +
  geom_sf(aes(fill = population)) +
   geom_label(aes(x = text_x, y = text_y, label = city), 
               size = 1.7, family = "HiraginoSans-W3") +
  scale_fill_distiller(name = "人口",
                       palette = "Greens", direction = 1) +
  theme_map(base_family = "HiraginoSans-W3") +
  theme(legend.position = c(.8, .05),
        legend.title = element_text(size = 10), 
        legend.text = element_text(size = 5),
        legend.key.size = unit(0.5, "cm"),
        legend.key.width = unit(1,"cm")) +
  coord_sf(datum = NA) 

map_pop_tokyo_text

ggsave("fig/map_pop_tokyo_text.png", map_pop_tokyo_text, width = 13, height = 13)

library(tidyverse)
library(jpndistrict)
library(sf)
library(ggthemes)
library(rmarkdown)

df_dist_map <- sf::st_read("shp/senkyoku289polygon.shp",
                       options = "ENCODING=CP932", 
                       stringsAsFactors = FALSE) 

head(df_dist_map) %>% 
  paged_table()

df_hr <- read_csv("https://raw.githubusercontent.com/yukiyanai/quant-methods-R/master/data_fixed/hr96-17.csv") 

head(df_hr) %>% 
  paged_table()

df_pref_en <- jpndistrict::jpnprefs %>% 
    mutate(
        prefecture_en = str_remove_all(prefecture_en, c("-ken" = "",  #-kenken を削除 
                                                        "-to" = "",   #-to を削除 
                                                        "-fu" = "")), #-fu を削除 
        prefecture_en = str_to_lower(prefecture_en),
        prefecture = str_remove_all(prefecture, c("県" = "",  # 県を削除  
                                                  "都" = "",  # 都を削除 
                                                  "府" = "")) # 府を削除 
    ) %>% 
    select(prefecture, ku = prefecture_en) # 変数を 2 つに絞る  

df_pref_en %>%
  paged_table()

df_hr17 <- df_hr %>% 
    left_join(df_pref_en, "ku") %>% 
    filter(year == 2017) %>% 
    mutate(
        district = str_c(prefecture, kun, "区") # 県名 + 数字 + 区
    ) 

df_hr17 %>% 
  paged_table()

df_hr17_age <- df_hr17 %>% 
    group_by(district) %>% 
    summarise(age_mean = mean(age, na.rm = T))

head(df_hr17_age)

head(df_dist_map)

names(df_dist_map)[2] <- "district"

names(df_dist_map)

df_dist_map_age <- df_dist_map %>% 
    left_join(df_hr17_age, by = "district") %>% 
    st_as_sf()

map_dist_sizuoka_age <- df_dist_map_age %>% 
    filter(str_detect(district, "静岡")) %>% 
    ggplot() +
    geom_sf(aes(fill = age_mean)) +
    scale_fill_distiller(name = "候補者の年齢の平均",
                         palette = "YlOrRd", direction = 1) +
    theme_map(base_family = "HiraginoSans-W3") +
    theme(legend.position = c(.1, -.1),
          legend.direction = "horizontal",
          legend.title = element_text(size = 15), 
          legend.text = element_text(size = 15),
          legend.key.size = unit(1, "cm"),
          legend.key.width = unit(3,"cm")) +
    coord_sf(datum = NA) 

map_dist_sizuoka_age

ggsave("fig/map_dist_sizuoka_age.png", map_dist_sizuoka_age,
       width = 10, height = 10)

map_dist_sizuoka_text <- df_dist_map_age %>% 
    filter(str_detect(district, "静岡")) %>% 
    mutate(
        text_x = map_dbl(geometry, ~st_centroid(.x)[[1]]),
        text_y = map_dbl(geometry, ~st_centroid(.x)[[2]]),
        district = str_remove_all(district, "静岡")
        ) %>% 
    ggplot() +
    geom_sf(aes(fill = age_mean)) +
    geom_label(aes(x = text_x, y = text_y, label = district), 
               size = 5, family = "HiraginoSans-W3") +
    scale_fill_distiller(name = "候補者の年齢の平均",
                         palette = "YlOrRd", direction = 1) +
    theme_map(base_family = "HiraginoSans-W3") +
    theme(legend.position = c(.1, -.1),
          legend.direction = "horizontal",
          legend.title = element_text(size = 15), 
          legend.text = element_text(size = 15),
          legend.key.size = unit(1, "cm"),
          legend.key.width = unit(3,"cm")) +
    coord_sf(datum = NA) 

map_dist_sizuoka_text

ggsave("fig/map_dist_sizuoka_text.png",map_dist_sizuoka_text,
       width = 10, height = 10)

LDP <- read_csv("hr09_14_ldp_seatshare.csv",
                         locale = locale(encoding = "cp932"))  

DT::datatable(LDP)

# year変数を factor に変換する
LDP$year <- factor(LDP$year) 

# order_seqという変数を加えて LDP_reordered というデータフレーム名をつける
LDP_reordered <- LDP %>%
  arrange(year, ldp_ratio) %>%
  mutate(order_seq = c(1:47, rep(0, 47*2))) 

ggplot(LDP_reordered, aes(x=ldp_ratio, y=reorder(pref, order_seq))) + 
  geom_segment(aes(yend=pref),
               xend=0, colour="grey50") +
  geom_point(size=2,
             aes(colour=year)) +
  scale_colour_brewer(palette="Set1", limits=c("2009", "2012", "2014"),
                      guide=FALSE) +
  theme(panel.grid.major.y =
          element_blank()) +
  facet_grid(~ year,
             scales="free_y", space="free_y")

## データセットを読み込む
LDP <- read_csv("data/hr09-14_ldp_seatshare.csv")

# year変数を factor に変換する
LDP$year <- factor(LDP$year)

LDP_reordered <- LDP %>%
  arrange(year, ldp_ratio) %>%
  mutate(order_seq = c(1:47, rep(0, 47*2)))

ggplot(LDP_reordered, aes(x = reorder(pref, order_seq), y = ldp_ratio,  fill = year)) +
  geom_bar(stat = "identity") +
  facet_grid(~ year, scales = "free_x") +
  theme(legend.position = "none") +
  labs(x = "Prefecture", y = "LDP's seat share (%)") +
  coord_flip()