Skip to content
Tutorials Center
My learning

Topics

Lesson 11 of 42 ยท Basic Charts

Area Charts and Time Series

Plot time series in Plotly: area and stacked area charts, percent areas, filling between lines, resampling and moving averages with pandas, range sliders and range selector buttons.

An area chart is a line chart with the space under the line filled in. The fill gives the eye a sense of volume, which is why area charts suit quantities that accumulate or add up: visitors, revenue, energy use. Stacked areas show how several series add up to a total and how the mix changes over time.

This lesson also collects the everyday tools for time series: preparing dates, smoothing with pandas, and the interactive range slider and range selector buttons that make long series easy to explore.

Area charts

An area trace is a scatter trace with a fill. px.area creates it, and color splits the data into stacked areas:

Python Stacked and percent areas

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Each continent became a trace in the same stackgroup, so plotly.js piles the values up instead of overlapping them. The printed fill is empty because a stackgroup already implies it: plotly.js fills the first trace down to zero and every later one up to the trace below, which is how a stack is drawn. groupnorm="percent" rescales every x position to 100%, so you see the mix instead of the total.

The same thing by hand

With graph objects you set stackgroup and fill yourself. A chart in plotly.js:

Stacked areas with plotly.js

Without stackgroup the areas would overlap and hide each other; with fill: "tozeroy" on each trace you can instead draw overlapping translucent areas (set opacity or fillcolor with an alpha value). Another useful fill is fill: "tonexty" between two lines, which shades a band such as a confidence interval.

Preparing time series with pandas

Before plotting a time series, make sure the date column is a real datetime, sorted, and (if needed) aggregated to the right resolution. Pandas does all of it:

Python Resample and smooth a time series

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Plotting the raw data in a light color and the smoothed trend in a strong color is the standard way to show both noise and signal. resample("W") aggregates days into weeks (use "ME" for month ends, "QE" for quarters). The first six values of a 7-day rolling mean are missing, which Plotly draws as a gap at the start.

Range slider and range selector

Time series are often long, and the interesting part may be a small window. Two built-in tools help. The range slider is a mini-overview below the chart with draggable handles. The range selector is a set of buttons ("1m", "6m", "YTD", "all") that jump to a window.

Range slider and selector buttons

In Python the same options are fig.update_xaxes(rangeslider_visible=True, rangeselector=dict(buttons=[...])). A detailed treatment, including hiding weekends with rangebreaks, is in Axes, Ticks and Scales.

Tips for time series charts

  • Use a line for continuous values and bars for totals per period (per month, per week).
  • Use markers only when points are few. Hundreds of markers turn a line into noise.
  • Keep the same y scale when you compare periods, or label clearly if it changes.
  • Add annotations for events (a launch, an outage) to explain bumps; the annotations lesson covers this.
  • For finance, candlestick and OHLC charts show open, high, low and close.

Check your understanding

Quick check ยท 5 questions

of 5 answered You scored / 5 ยท perfect!

  1. 1. What property makes several area traces pile up instead of overlap?

    Traces that share a stackgroup are stacked on top of each other. (barmode is for bars.)
  2. 2. What does groupnorm="percent" do for stacked areas?

    You see each series' share instead of absolute values.
  3. 3. Which pandas method turns daily data into weekly totals?

    resample groups a datetime index into periods and aggregates them.
  4. 4. What is the range selector?

    The mini-chart with handles is the range slider; the selector is the button row.
  5. 5. Why plot raw data in light grey and a moving average in a strong color?

    De-emphasizing the noisy series keeps context without hiding the signal.

Summary

  • Area charts are scatter traces with fill; px.area and stackgroup stack them, groupnorm="percent" shows shares.
  • Prepare time series with pd.to_datetime, sorting, resample and rolling.
  • Raw data plus a smoothed line shows noise and trend together.
  • rangeslider and rangeselector make long series explorable.

Next: Sunburst, Treemap and Icicle Charts