Boxplot chart

Boxplot charts give a good graphical image of the concentration of data. In Siren Investigate, you can create a boxplot chart from the numeric data in a set of Elasticsearch documents.

A boxplot is constructed from the following five values:

  • The minimum value;

  • The first quartile;

  • The median;

  • The third quartile; and

  • The maximum value.

These values are used for comparison purposes: You can compare how close other data values are to them.

For example, you might want to chart and analyze the amount of money that was invested in a company in different currencies.

Boxplot

Creating a boxplot chart

Before you begin

To create a boxplot chart, the following data metrics are required:

  • One Percentiles metric that contains three defined percentiles:

    • Bottom percentile (usually around 25%)

    • Median (usually around 50%)

    • Top percentile (usually around 75%)

  • One Min metric

  • One Max metric

The following data metrics are optional:

  • One Count metric

  • One Aggregation (bucket)

  • One Sub Aggregation (bucket)

Procedure

  1. In the sidebar navigation menu, click Visualize→Create visualization.

  2. Under the Charts section, select Boxplot Chart from the list of available visualizations.

  3. Select an entity table or search from the list.

  4. On the Data tab, select the metrics type Parameter.

  5. In the Aggregation dropdown list, select Percentiles.

  6. Select a Field from the dropdown list and add or remove any percentage values as required.

    A typical use case contains the 25%, 50%, and 75% values. If you want to add a percentage value that is not in the list, click Add.

  7. Click Add metrics and select a parameter for the metric aggregation Min.

  8. Select the Field from the dropdown list.

  9. Click Add metrics and select a parameter for the metric aggregation Max.

  10. Select the Field from the dropdown list.

  11. In the buckets table window, click Grouping Aggregation.

  12. Select an aggregation and specify the values. For example, select Date Range and specify the start and end dates.

  13. If your chart is complete and you don’t want to make any formatting changes, click Apply changes and save the visualization to a dashboard.

You can also add the properties in JSON format by clicking Advanced.

Formatting a boxplot chart

On the Options tab, you can specify the following formatting settings for the chart:

  • Y Axis Text: Specify a label for the Y axis.

  • X Axis Text: Specify a label for the X axis.

  • Show values: Select this check box to display the value next to its box.

  • Rotate X Axis Labels: You can set the label of the X axis to vertical, horizontal, or angled.

  • Y Axis Scale: You can set the scale of the Y axis to linear, log, or square root.

  • Restrict Y axis MAX: Select this check box to restrict the domain of the Y axis to a maximum value.

  • Restrict Y axis MIN: Select this check box to restrict the domain of the Y axis to a minimum value.

When the chart is complete, click Apply changes and save the visualization to a dashboard.