Histogram Series

Histograms show the frequency distribution of continuous data. They are a good choice for when the data is larger than could be plotted on a bar chart and can be used to find underlying trends in continuous data.

Simple histogram

Histograms require at least one numeric attribute in the data to be specified using the xKey property. Data will be distributed into bins according to the xKey values.

The simplest configuration for a Histogram Series is shown below:

series: [{ type: 'histogram' xKey: 'age' }]

Bin Count

By default the histogram will split the x domain of the data into around ten regular-width bins, although the exact number generated will vary so that the chart can find round values for the bin boundaries. This is similar to how giving a number of ticks to an axis does not guarantee that exact number of ticks. The number of bins to aim for can be overridden by setting the binCount property on a histogram series.

Given enough data, charts with more bins are able to more precisely illustrate underlying trends, but are also more sensitive to random noise.

series: [{ type: 'histogram' xKey: 'age', binCount: 20 }]

Irregular intervals

Rather than specify the number of bins, for cases where you know exactly which bins you wish to split your x-axis values into, it is possible to explicitly give the start and end values for each bin. This is given using the bins property, and the value should be an array of arrays where each inner array contains the start and end value of a bin. In the example below, the data from the race is split into irregular age categories.

For histogram charts with irregular bins it is usual for the area of the bar, rather than its height, to visually represent the value of each bin. In this way the shape of the underlying curve is maintained over irregular intervals. The areaPlot property should be set to true to enable this mode.

series: [{ type: 'histogram' xKey: 'age', areaPlot: true, bins: [[16, 18], [18, 21], [21, 25], [25, 40]] }]

Note that if you give the bins property you should not also give binCount, but if both are present bins takes precedence.

XY Histogram

The histograms shown above all contain a single xKey with it's frequency plotted on the y axis. However it is sometimes useful to visualise an xKey and yKey using a Histogram.

When using XY Histograms it is also useful to control how bins are aggregated using the aggregation series property. The following sections contrast the sum and mean aggregation functions.

Summing Bins

To show the summing of one column or attribute for each of the bins. When a yKey is given the default behaviour is to plot a total of the yKey values. The kind of aggregation to use is controlled by the series.aggregation property.

series: [{ type: 'histogram' xKey: 'age', yKey: 'winnings', aggregation: 'sum' }]

Mean Bins

Showing frequencies or summing up the y-values isn't always the best way to show your data. For data that is not evenly distributed in x, but is relatively uniform in y, a sum plot xy histogram will tend to be dominated by the populations of the x-bins. In the above example you may notice that the prize money distribution very closely follows the age distribution, so that while potentially useful, the chart does not reveal any new trends in the data. In many cases, plotting the mean of a bin on the y-axis better illustrates an underlying trend in the data:

series: [{ type: 'histogram' xKey: 'age', yKey: 'time', yName: 'Average Time', aggregation: 'total' }]

API Reference

data
Required
The data to use when rendering the series. If this is not supplied, data must be set on the chart instead.
visibleWhether or not to display the series.
Default: true
showInLegendWhether or not to include the series in the legend.
Default: true
tooltipEnabledWhether or not to show tooltips when the series are hovered over.
Default: true
tooltipRendererFunction used to create the content for tooltips.function (params: IParams): string; interface IParams { datum: any; title?: string; color?: string; xKey: string; xName?: string; yKey: string; yName?: string; sizeKey?: string; sizeName?: string; labelKey?: string; labelName?: string; }
xKey
Required
The key to use to retrieve x-values from the data.
xNameA human-readable description of the x-values.
yKey
Required
The key to use to retrieve y-values from the data.
yNameA human-readable description of the y-values.
binCountThe number of bins to try to split the x axis into. Clashes with the bins setting.
binsSet the bins explicitly. The bins need not be of equal width. Clashes with the binCount setting.
aggregationDictates how the bins are aggregated. If set to 'sum', the value shown for the bins will be the total of the yKey values. If set to 'mean', it will display the average yKey value of the bin
Default: 'sum'
Options: 'sum', 'mean'
areaPlotFor variable width bins, if true the histogram will represent the aggregated yKey values using the area of the bar. Otherwise, the height of the var represents the value as per a normal bar chart. This is useful for keeping an undistorted curve displayed when using variable-width bins
Default: 'false'
highlightStyleConfiguration for the highlighting used when the markers are hovered over. See highlightStyle for more details about this configuration object.
fillThe colour of the fill for the histogram bars.
Default: '#f3622d'
fillOpacityThe opacity of the fill for the histogram bars.
Default: 1
strokeThe colour of the stroke for the histogram bars.
Default: '#aa4520'
strokeOpacityThe opacity of the stroke for the histogram bars.
Default: 1
strokeWidthThe width in pixels of the stroke for the histogram bars.
Default: 1
listenersA map of event names to event listeners. See listeners for more details about this configuration object.

highlightStyle

Configuration for the highlighting used when the markers are hovered over.

fillThe fill colour of the markers when hovered over.
Default: 'yellow'
strokeThe colour of the stroke around the markers when hovered over.

listeners

A map of event names to event listeners.

nodeClickThe listener to call when a histogram bar is clicked.function (params: IParams): any; interface IParams { type: 'nodeClick'; series: HistogramSeries; datum: any; xKey: string; }

Next Up

Continue to the next section to learn about layout.