Powered by AppSignal & Oban Pro
Would you like to see your link here? Contact us
Notesclub
Sign in with GitHub
created by hec & contributors
terms privacy

チョコレート統計

chocolate_statistics.livemd

avatar
Ryo Wakabayashi
avatar
avatar
More notebooks
0

チョコレート統計

Mix.install([
  {:explorer, "~> 0.8"},
  {:kino, "~> 0.12"},
  {:kino_vega_lite, "~> 0.1"},
  {:req, "~> 0.3"}
])

出典

以下のデータを加工して作成

総務省統計局ホームページ

家計調査（家計収支編）　時系列データ（二人以上の世帯）

  • 月 全品目（2015年改定）
  • 月 全品目（2020年改定）

（2023年2月20日に利用）

準備

alias Explorer.DataFrame
alias Explorer.Series
require Explorer.DataFrame

データ取得

household_df =
  "https://raw.githubusercontent.com/RyoWakabayashi/elixir-learning/main/livebooks/explorer/%E5%AE%B6%E8%A8%88%E6%94%AF%E5%87%BA%E7%B5%B1%E8%A8%88_%E5%93%81%E7%9B%AE%E5%B9%B4%E6%9C%88%E5%88%A5.csv"
  |> Req.get!()
  |> then(&DataFrame.load_csv!(&1.body))

Kino.DataTable.new(household_df)

月次推移

choco_df = DataFrame.filter(household_df, 符号 == "352")

Kino.DataTable.new(choco_df)
month = Series.to_list(choco_df["年月"])
expenses = Series.to_list(choco_df["支出金額"])

VegaLite.new(width: 700, title: "チョコレート支出金額推移")
|> VegaLite.data_from_values(x: month, y: expenses)
|> VegaLite.mark(:line, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :temporal, title: "年月")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "支出金額")
2015..2022
|> Enum.map(fn year ->
  df = DataFrame.filter(choco_df, 年 == ^year)

  month = Series.to_list(df["月"])
  expenses = Series.to_list(df["支出金額"])

  graph =
    VegaLite.new(width: 700, title: "チョコレート支出金額推移")
    |> VegaLite.data_from_values(x: month, y: expenses)
    |> VegaLite.mark(:line, tooltip: true)
    |> VegaLite.encode_field(:x, "x", type: :ordinal, title: "月")
    |> VegaLite.encode_field(:y, "y",
      type: :quantitative,
      title: "支出金額",
      scale: [domain: [0, 1500]]
    )

  {year, graph}
end)
|> Kino.Layout.tabs()

年次推移

year_choco_df =
  choco_df
  |> DataFrame.group_by("年")
  |> DataFrame.summarise(
    最小: min(支出金額),
    最大: max(支出金額),
    平均: mean(支出金額),
    合計: sum(支出金額)
  )

Kino.DataTable.new(year_choco_df)
year = Series.to_list(year_choco_df["年"])
expenses = Series.to_list(year_choco_df["合計"])

VegaLite.new(width: 700, title: "チョコレート年間支出金額推移")
|> VegaLite.data_from_values(x: year, y: expenses)
|> VegaLite.mark(:line)
|> VegaLite.encode_field(:x, "x", type: :ordinal, title: "年")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "合計支出金額")
pivot_table =
  DataFrame.pivot_longer(
    year_choco_df,
    ["最小", "最大", "平均"],
    names_to: "統計項目",
    values_to: "統計値"
  )

Kino.DataTable.new(pivot_table)
year = Series.to_list(pivot_table["年"])
items = Series.to_list(pivot_table["統計項目"])
values = Series.to_list(pivot_table["統計値"])

VegaLite.new(width: 700, title: "チョコレート年間支出金額推移")
|> VegaLite.data_from_values(x: year, items: items, values: values)
|> VegaLite.mark(:line)
|> VegaLite.encode_field(:x, "x", type: :ordinal, title: "年")
|> VegaLite.encode_field(:y, "values", type: :quantitative, title: "支出金額")
|> VegaLite.encode_field(:color, "items")

菓子類内訳

snack_df =
  DataFrame.filter(
    household_df,
    大分類 == "1" and
      中分類 == "8" and
      符号 != "-"
  )

Kino.DataTable.new(snack_df)
latest_snack_df = DataFrame.filter(snack_df, 年 == 2022 and 月 == 12)

items = Series.to_list(latest_snack_df["品目分類"])
expenses = Series.to_list(latest_snack_df["支出金額"])

VegaLite.new(width: 700, title: "菓子類支出金額")
|> VegaLite.data_from_values(x: items, y: expenses)
|> VegaLite.mark(:bar, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :nominal, title: "品目分類", sort: "-y")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "支出金額")
VegaLite.new(title: "菓子類支出金額")
|> VegaLite.data_from_values(x: items, y: expenses)
|> VegaLite.mark(:arc, inner_radius: 50, tooltip: true)
|> VegaLite.encode_field(:color, "x", type: :nominal, title: "品目分類")
|> VegaLite.encode_field(:theta, "y", type: :quantitative, title: "支出金額")
total_snack_df =
  DataFrame.filter(
    household_df,
    大分類 == "1" and
      中分類 == "8" and
      符号 == "-"
  )

Kino.DataTable.new(total_snack_df)
month = Series.to_list(total_snack_df["年月"])
expenses = Series.to_list(total_snack_df["支出金額"])

VegaLite.new(width: 700, title: "菓子類支出金額推移")
|> VegaLite.data_from_values(x: month, y: expenses)
|> VegaLite.mark(:line, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :temporal, title: "年月")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "支出金額")
choco_snack_df = DataFrame.filter(household_df, 符号 == "353")

ratio_df =
  total_snack_df[["年月", "支出金額"]]
  |> DataFrame.join(
    choco_df[["年月", "支出金額"]],
    on: ["年月"]
  )
  |> DataFrame.rename(["年月", "菓子類合計", "チョコレート"])
  |> DataFrame.join(
    choco_snack_df[["年月", "支出金額"]],
    on: ["年月"]
  )
  |> DataFrame.rename(["年月", "菓子類合計", "チョコレート", "チョコレート菓子"])
  |> DataFrame.mutate(
    チョコレート率: チョコレート / 菓子類合計,
    チョコレート菓子率: チョコレート菓子 / 菓子類合計
  )
  |> DataFrame.mutate(チョコレート類率: チョコレート率 + チョコレート菓子率)
  |> DataFrame.pivot_longer(
    &String.ends_with?(&1, "率"),
    names_to: "品目",
    values_to: "占有率"
  )

Kino.DataTable.new(ratio_df)
month = Series.to_list(ratio_df["年月"])
items = Series.to_list(ratio_df["品目"])
ratio = Series.to_list(ratio_df["占有率"])

VegaLite.new(width: 600, title: "占有率推移")
|> VegaLite.data_from_values(x: month, y: ratio, color: items)
|> VegaLite.mark(:line, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :temporal, title: "年月")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "占有率")
|> VegaLite.encode_field(:color, "color")
year_total_snack_df =
  total_snack_df
  |> DataFrame.group_by("年")
  |> DataFrame.summarise(合計: sum(支出金額))

Kino.DataTable.new(year_total_snack_df)
year = Series.to_list(year_total_snack_df["年"])
expenses = Series.to_list(year_total_snack_df["合計"])

VegaLite.new(width: 700, title: "菓子類年間支出金額推移")
|> VegaLite.data_from_values(x: year, y: expenses)
|> VegaLite.mark(:line, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :ordinal, title: "年")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "合計支出金額")
year_choco_snack_df =
  choco_snack_df
  |> DataFrame.group_by("年")
  |> DataFrame.summarise(合計: sum(支出金額))

year_ratio_df =
  year_total_snack_df[["年", "合計"]]
  |> DataFrame.join(
    year_choco_df[["年", "合計"]],
    on: ["年"]
  )
  |> DataFrame.rename(["年", "菓子類合計", "チョコレート"])
  |> DataFrame.join(
    year_choco_snack_df[["年", "合計"]],
    on: ["年"]
  )
  |> DataFrame.rename(["年", "菓子類合計", "チョコレート", "チョコレート菓子"])
  |> DataFrame.mutate(
    チョコレート率: チョコレート / 菓子類合計,
    チョコレート菓子率: チョコレート菓子 / 菓子類合計
  )
  |> DataFrame.mutate(チョコレート類率: チョコレート率 + チョコレート菓子率)
  |> DataFrame.pivot_longer(
    &String.ends_with?(&1, "率"),
    names_to: "品目",
    values_to: "占有率"
  )

Kino.DataTable.new(year_ratio_df)
year = Series.to_list(year_ratio_df["年"])
items = Series.to_list(year_ratio_df["品目"])
ratio = Series.to_list(year_ratio_df["占有率"])

VegaLite.new(width: 600, title: "占有率推移")
|> VegaLite.data_from_values(x: year, y: ratio, color: items)
|> VegaLite.mark(:line, tooltip: true)
|> VegaLite.encode_field(:x, "x", type: :ordinal, title: "年")
|> VegaLite.encode_field(:y, "y", type: :quantitative, title: "占有率")
|> VegaLite.encode_field(:color, "color")
Back