Area range chart (advanced)

import easychart

# data is pd.Series of electric and gas utility industrial production
data = easychart.datasets.load("electricity")

# detrend the data naively
growth = (data / data.shift(12) - 1).fillna(0).apply(lambda x: (1 + x) ** (1 / 12) - 1)
data = data.divide((1 + growth.ewm(alpha=0.25).mean()).cumprod())
# slice the data since 1975
data = data["1975-01-01":]
# rebase the data to average 100
data = 100 * data / data.mean()

# create a pivot (month, year)
pivot = data.groupby([data.index.month, data.index.year]).last().unstack()
pivot.index = [
    "Jan",
    "Feb",
    "Mar",
    "Apr",
    "May",
    "Jun",
    "July",
    "Aug",
    "Sep",
    "Oct",
    "Nov",
    "Dec",
]

# compute different levels of historical norms
for decile in [0, 0.05, 0.25, 0.5, 0.75, 0.95, 1]:
    pivot[f"{decile:.0%}"] = pivot.iloc[:, :-1].quantile(decile, axis=1)

# create the chart
chart = easychart.new(tooltip="shared", title="US gas and electricity production")
chart.subtitle = "Detrended (100 = 1975-2020 average)"
chart.categories = pivot.index
chart.marker = False
chart.plot(
    zip(pivot["5%"], pivot["95%"]),
    name="10%-90% range",
    type="arearange",
    color="#eeeeee",
)
chart.plot(
    zip(pivot["25%"], pivot["75%"]),
    name="25%-75% range",
    type="arearange",
    color="#cccccc",
)
chart.plot(pivot["0%"], name="min", dashstyle="dot", color="black")
chart.plot(pivot["100%"], name="max", dashstyle="dot", color="black")
chart.plot(pivot[2019], name="2019")
chart.plot(pivot[2020], name="2020")
chart