You can choose from among several statistical methods to identify and adjust outliers and fill in missing values.
To select an outlier detection method:
In the Data Attributes panel, click View Screened Data.
The Historical Data - Data Screening dialog opens.
In Historical Data — Data Screening, click Screening Options.
The Data Screening Options dialog opens.
Select a detection method and enter an associated threshold value.
You can select outliers using the mean and standard deviation, the median and median absolute deviation (MAD), or the median and interquartile deviation (IQD). For a description of each method, see the Predictor sections of the Oracle Crystal Ball Reference and Examples Guide. The default is Mean and Standard Deviation with a standard deviation of 3.
To select a method for adjusting outliers and filling in missing values:
Display the Data Screening Options dialog as described in steps 1 and 2 above.
Cubic spline interpolation calculates a smooth, continuous curve that passes through each data point. It evaluates the entire data set.
Neighbor interpolation examines values on each side of the value to be adjusted or filled in and calculates that value based on the mean or median of the specified neighbors.
For more information about each method, see the Predictor sections of the Oracle Crystal Ball Reference and Examples Guide.
If you select Neighbor interpolation, indicate the number of neighbors to evaluate on each side of the target value and select a statistic.