Adding Forecasts in Visuals

The built-in machine learning algorithm lets you easily forecast key data metrics. For example, you can forecast printer throughput to see if you can meet your goal by the end of the year. Anomalies in the data are excluded so they do not influence the forecast. You can also create interactive what-if analyses to determine the growth trajectory that you need to meet goals.
To add forecasts:
  1. Click Dashboards tab, the Dashboards tab.
  2. Click Dashboard Designer, in the bottom-left corner.
  3. On the Analyses dialog, click Options button, the Options button, next to an analysis and select Edit.
  4. On the analysis page, click Menu options button, the Menu options button, in the corner of a visual that displays data over time and select Add forecast.
    The machine learning algorithm automatically analyzes the historical data and displays a graphical forecast for the next 14 periods. Forecast properties apply to all the data metrics in the visual.

    You can add forecasts only in visuals that use a single date field and up to three measure fields.

  5. In the Forecast properties panel, customize the forecast settings.
    1. Under Forecast length, set the number of periods forward to forecast or the number of periods backward to analyze for patterns.
    2. Under Prediction interval, set the estimated range for the forecast to change how wide the band of possibility is around the predicted line.
      The prediction interval is an estimate of an interval in which future observations can fall, with a certain probability, based on what has already been observed.
    3. Under Seasonality, set the number of time periods involved in the predictable seasonal pattern of data.
      Seasonality is a characteristic of a time series in which the data experiences regular and predictable changes that recur every calendar year.
    4. Under Forecast boundaries, set a minimum and maximum forecast value to prevent forecast values from going above or below a specified threshold. For example, you can stop the forecasted values from ever going below zero.
  6. Click Apply.
    If your forecast contains multiple data metrics, you can isolate one of the forecasts by clicking anywhere inside the forecast band. Click the isolated forecast band again to display all the forecasts.
  7. To analyze what-if scenarios, right-click a forecasted data point from the forecast band and select What-if analysis from the menu.
    The What-if analysis option is only available for visuals that contain only one data metric.
  8. Customize the settings from the What-if analysis panel.
    1. Under Scenario, specify if you want to set a target for a date or set a target for a time range.
    2. Under Date, enter the target date or the start and end dates of the target time range.
    3. Under Target, set a target value for the data metric.
  9. Click Apply.
    A new forecast adjusted for the target is displayed alongside the original forecast. The what-if analysis is represented on the visual as a dot on the data metric line. You can hover over the data points on the forecasting line to see the details.

    To interact with or remove a what-if analysis, click the dot on the data metric line. To create other what-if scenarios, close the what-if analysis before choosing a new point on the line.

    Note: What-if scenarios are only supported in analyses. They are not available in published dashboards.

To edit or remove a forecast, click Menu options button, the Menu options button, in the top-corner of the visual and select Edit forecast. Update the forecast settings or click Remove to remove the forecast.

Parent topic: Using Insights, Anomaly Detection, and Forecasts