Scenario Definition (Temporal Causal Model Scenarios)

Settings for defining a scenario depend on whether your data are column-based or multidimensional.

Root series

Specifies the root series for the scenario. Each scenario is based on a single root series. For column-based data, you select the field that defines the root series. For multidimensional data, you specify the root series by adding an entry to the grid for the metric field that contains the series. You then specify the values of the dimension fields that define the root series. The following apply to specifying the dimension values:

• You can enter the value for each dimension field directly into the grid or you can select from the list of available dimension values. To select from the list of available dimension values, click the ellipsis button in the cell for the dimension that you want. This action opens the Select Dimension Value subdialog.
• You can search the list of dimension values, on the Select Dimension Value subdialog, by clicking the binoculars icon and specifying a search term. Spaces are treated as part of the search term. Asterisks (*) in the search term do not indicate wildcard characters.

Specify affected targets

Use this option when you know specific targets that are affected by the root series and you want to investigate the effects on those targets only. By default, targets that are affected by the root series are automatically determined. You can specify the breadth of the series that are affected by the scenario with settings on the Options tab.

For column-based data, select the targets that you want. For multidimensional data, you specify target series by adding an entry to the grid for the target metric field that contains the series. By default, all series that are contained in the specified metric field are included. You can customize the set of included series by customizing the included values for one or more of the dimension fields. To customize the dimension values that are included, click the ellipsis button for the dimension that you want. This action opens the Select Dimension Values dialog.

The Series Count column (for multidimensional data) displays the number of sets of dimension values that are currently specified for the associated target metric. The displayed value can be larger than the actual number of affected target series (one series per set). This condition occurs when some of the specified combinations of dimension values do not correspond to series contained by the associated target metric.

Scenario ID

Each scenario must have a unique identifier. The identifier is displayed in output that is associated with the scenario. There are no restrictions, other than uniqueness, on the value of the identifier.

Specify scenario values for root series

Use this option to specify explicit values for the root series in the scenario period. You must specify a numeric value for each time interval that is listed in the grid. You can obtain the values of the root series (actual or forecasted) for each interval in the scenario period by clicking Read, Forecast, or Read\Forecast.

Specify expression for scenario values for root series

You can define an expression for computing the values of the root series in the scenario period. You can enter the expression directly or click the calculator button and create the expression from the Scenario Values Expression Builder.

• The expression can contain any target or input in the model system.
• When the scenario period extends beyond the existing data, the expression is applied to forecasted values of the fields in the expression.
• For multidimensional data, each field in the expression specifies a time series that is defined by the field and the dimension values that were specified for the root metric. It is those time series that are used to evaluate the expression.

As an example, assume that the root field is advertising and the expression is advertising*1.2. The values for advertising that are used in the scenario represent a 20 percent increase over the existing values.