Regression Line

Example

Illustration

Summary of settings

  • Type of chart: Scatter
  • Mapping of y attribute: Profit
  • Mapping of x attribute: Size
  • Cardinality: City
  • Overlays: polynomial 2 order regression line

Step-by-step instruction

  1. Add new Chart Widget

  2. Select Scatter

  3. If there are multiple cubes connected to your application, you must select the cube you want for the chart.

  4. To map the x attribute, select a measure or a formula for the horizontal axis

  5. Map the y attribute in the same way as was done for the x attribute.

  6. The chart displays with one point, the total for the measure selected on the x axis plotted against the total for the measure selected on the y axis.

  7. To select Members you would like to plot ( members of the dimension City for example) , click Options in the Legend area of the chart:

    Illustration

    In the Options popup you can change various attributes, in this case map Cardinality to City:

    Illustration

    Now, six points display on the scatter plot, because there are six members in the City dimension with Size and Profit linked to them:

    Illustration

    Use filters to limit the scope of data in the view Filters

  8. Now go back to Options (in the Legend) and switch to the Overlays tab to add a new overlay, in this case, a regression line:

    Illustration

    On mouse-hover, you can see the estimated parameters:

    Illustration

The chart is complete:

Illustration

Back to Legacy Chart Gallery

Appendix

The appendix contains code snippets for advanced users.

MDX:

SELECT
  {
    [Measures].[Profit],
    [Measures].[Size]
  } ON COLUMNS,
  NON EMPTY [Geography].[City].[City].Members ON ROWS
  FROM [EquityDerivativesCube]

JSON:

{
  "configurations": [
    {
      "handlers": {
        /* ... */
      },
      "type": "scatter",
      "mapping": {
        "x": {
          "from": "[Measures].[Size]"
        },
        "y": {
          "from": "[Measures].[Profit]"
        },
        "cardinality": {
          "from": ["[Geography].[City].[City]"]
        }
      },
      "overlays": [
        {
          "key": "regression",
          "args": {
            "type": "polynomial"
          }
        }
      ]
    }
  ]
}