Smart Sensitivity Analysis

By default, @RISK uses “Smart Sensitivity Analysis”. This powerful technique pre-screens inputs, and removes those that are not connected, via a formulaic relationship to the output. Inputs that have no link to an output cell are removed from the sensitivity analysis, thus avoiding spurious results that arise simply from random variations. In the Sensitivity Analysis window, these unrelated inputs are indicated by “N/A”.

There are some situations where disabling Smart Sensitivity Analysis makes sense (See Sampling Settings under the Simulation Settings section for more information on how to disable Smart Sensitivity Analysis):

• Smart Sensitivity Analysis requires the Excel precedence tree to be scanned. This can add a significant amount of runtime for very large models. This increased runtime might outweigh the desire to avoid spurious correlations.

• The model contains a macro or DLL that performs calculations involving @RISK input values in cells that have no relationship (via workbook formulas) with the output. This macro or DLL then returns a result to a cell that is used in calculating the output values. In this case, Smart Sensitivity Analysis should be disabled. To avoid situations like this, we recommend the use of macro functions (UDFs) that explicitly reference all used input cells in their argument lists.

• @RISK is unable to scan through certain Excel functions to find the input/output connection. The INDIRECT and OFFSET functions are two examples. @RISK presents warnings in these circumstances, but disabling Smart Sensitivity Analysis will suppress those messages.