Change in Output Statistic Analysis

In a change in output statistic analysis, samples for an input are grouped in a set of equal-sized bins (or “scenarios”), ranging from the input’s lowest value to its highest. A value for a statistic of the output (such as its mean) is calculated for the output values in the iterations associated with each bin. Inputs are ranked by the amount of positive/negative swing they caused for the output statistic.

This analysis is similar to @RISK’s Scenario Analysis, except that scenarios for each input are used instead of scenarios for an output. In the change in output statistic analysis:

  1. @RISK sorts all iterations in ascending order of that input.
  2. It divides the sorted iterations into a set number of bins. For example, with 2500 iterations and using ten (10) bins, the first bin would contain the 250 iterations with the 250 lowest values of this input; the second bin would contain the 250 iterations with the next lowest values of this input; and so on.
  3. @RISK computes a selected statistic of the output values in each bin (the mean, by default).

In a tornado chart, the lowest of the ten output statistics is the number at the left edge of this input’s bar, and the highest of the ten output statistics is the number at the right edge of the bar.

The key is that the inputs are first sorted in ascending order and binned in that order, and then an output statistics is computed for just the iterations in each bin.

In the Sensitivities window, inputs are ranked by the range between their highest statistic value for any bin and their lowest statistic value.

The output statistic for this analysis and the number of bins used for each input are configured through the Sensitivity Settings dialog of the Sensitivities window.