Correlations (simulation)

Input fields to be simulated are often known to be correlated--for example, height and weight. Correlations between inputs that will be simulated must be accounted for in order to ensure that the simulated values preserve those correlations.

Recalculate correlations when fitting. This choice specifies that correlations between simulated inputs are automatically calculated when fitting distributions to the active dataset through the Fit All or Fit actions on the Simulated Fields settings.

Do not recalculate correlations when fitting. Select this option if you want to manually specify correlations and prevent them from being overwritten when automatically fitting distributions to the active dataset. Values that are entered in the Correlations grid must be between -1 and 1. A value of 0 specifies that there is no correlation between the associated pair of inputs.

Reset. This resets all correlations to 0.

Use fitted multiway contingency table for inputs with a categorical distribution. For inputs with a categorical distribution, you can automatically compute a multiway contingency table from the active dataset that describes the associations between those inputs. The contingency table is then used when data are generated for those inputs. If you choose to save the simulation plan, the contingency table is saved in the plan file and is used when you run the plan.
  • Compute contingency table from the active dataset. If you are working with an existing simulation plan that contains a contingency table, you can recompute the contingency table from the active dataset. This action overrides the contingency table from the loaded plan file.
  • Use contingency table from loaded simulation plan. By default, when you load a simulation plan that contains a contingency table, the table from the plan is used. You can recompute the contingency table from the active dataset by selecting Compute contingency table from the active dataset.