# Adding Scenarios and Scenario Groups (Temporal Causal Model Scenarios)

The Scenarios tab specifies the scenarios that are to be run. To define scenarios, you must first define the scenario period by clicking Define Scenario Period. Scenarios and scenario groups (applies only to multidimensional data) are created by clicking the associated Add Scenario or Add Scenario Group button. By selecting a particular scenario or scenario group in the associated grid, you can edit it, make a copy of it, or delete it.

## Column-based data

The Root field column in the grid specifies the time series field whose values are replaced with the scenario values. The Scenario values column displays the specified scenario values in the order of earliest to latest. If the scenario values are defined by an expression, then the column displays the expression.

## Multidimensional data

- Individual Scenarios
- Each row in the Individual Scenarios grid specifies a time series whose values are replaced by the specified scenario values. The series is defined by the combination of the field that is specified in the Root Metric column and the specified value for each of the dimension fields. The content of the Scenario values column is the same as for column-based data.
- Scenario Groups
- A scenario group defines a set of scenarios that are based on a single root
metric field and multiple sets of dimension values. Each set of dimension values (one value
per dimension field), for the specified metric field, defines a time series. An individual
scenario is then generated for each such time series, whose values are then replaced by the
scenario values. Scenario values for a scenario group are specified by an expression, which is
then applied to each time series in the group.
The Series Count column displays the number of sets of dimension values that are associated with a scenario group. The displayed value can be larger than the actual number of time series that are associated with the scenario group (one series per set). This condition occurs when some of the specified combinations of dimension values do not correspond to series contained by the root metric for the group.

As an example of a scenario group, consider a metric field

`advertising`and two dimension fields`region`and`brand`. You can define a scenario group that is based on`advertising`as the root metric and that includes all combinations of`region`and`brand`. You might then specify`advertising*1.2`

as the expression to investigate the effect of increasing`advertising`by 20 percent for each of the time series that are associated with the`advertising`field. If there are 4 values of`region`and 2 values of`brand`, then there are 8 such time series and thus 8 scenarios defined by the group.