For greatest forecast accuracy, you should correlate related assumptions (Defining Correlations Between Assumptions). When you define a correlation, you assign a correlation coefficient, a number between -1.0 and +1.0 that measures the strength of the relationship. A positive value means that when one assumption is high, the other likely is high. A negative value means that the assumptions are inversely related; when one is high, the other likely is low.
You can use the Define Correlations feature of Crystal Ball to define correlations among assumptions in two ways:
Pairwise in List view, Correlating an Assumption with Others and Correlating Assumptions with Definitions in List View
Using a matrix, Correlating a Group of Assumptions with Each Other and Correlating Assumptions in Matrix View
Pairwise correlation definitions are applied directly to pairs of assumptions. Matrix correlation definitions are created in a block of cells in a dialog or workbook and applied to a group of assumptions. Both methods use the Define Correlations dialog, described in About the Define Correlations Dialog.
For correlation guidelines, see Guidelines for Correlating Assumptions.
A correlation matrix is created whenever two or more assumptions are correlated. Each assumption can belong to only one matrix. Noncorrelated assumptions can be added to the current matrix at any time. Both List view and Matrix view are views of the same matrix. For more information about correlation matrixes in Crystal Ball, see About Crystal Ball Correlation Matrixes.
Note: | Crystal Ball uses Spearman rank order correlation for all correlation computations to relate assumptions with different distribution types. For more information about Spearman correlations, see the “Statistical Definitions” chapter of the Oracle Crystal Ball Reference and Examples Guide. |