Simulation Fitting Node Settings Tab

Source node name. You can generate the name of the generated (or updated) Simulation Generate node automatically by selecting Auto. The automatically generated name is the name specified in the Simulation Fitting node if a custom name has been specified (or Sim Gen if no custom name has been specified in the Simulation Fitting node). Select Custom to specify a custom name in the adjoining text field. Unless the text field is edited, the default custom name is Sim Gen.

Fitting Options With these options you can specify how the distributions are fitted to the fields and how the fit of the distributions is assessed.
  • Number of cases to sample. This specifies the number of cases to use when fitting distributions to the fields in the data set. Select All cases to fit distributions to all of the records in the data. If your data set is very large, you might want to consider limiting the number of cases that are used for distribution fitting. Select Limit to first N cases to use only the first N cases. Click the arrows to specify the number of cases to use. Alternatively, you can use an upstream node to take a random sample of records for distribution fitting.
  • Goodness of fit criteria (continuous fields only). For continuous fields, select either the Anderson-Darling test or the Kolmogorov-Smirnoff test of goodness of fit to rank distributions when fitting distributions to the fields. The Anderson-Darling test is selected by default and is especially recommended when you want to ensure the best possible fit in the tail regions. Both statistics are calculated for every candidate distribution, but only the selected statistic is used to order the distributions and determine the best fitting distribution.
  • Bins (empirical distribution only). For continuous fields, the Empirical distribution is the cumulative distribution function of the historical data. It is the probability of each value, or range of values, and is derived directly from the data. You can specify the number of bins that are used for calculating the Empirical distribution for continuous fields by clicking the arrows. The default is 100 and the maximum is 1000.
  • Weight field (optional). If your data set contains a weight field, click the field picker icon and select the weight field from the list. The weight field is then excluded from the distribution fitting process. The list shows all fields in the data set that have the measurement level continuous. You can select only one weight field.