Day: September 17, 2010

Capturing Dependencies with Correlations – Part III: Simulated Correlations

We’ve seen how you can define correlation coefficients in your models to describe the relationship between two or more variables. But during an actual risk analysis simulation, how can you be sure that the values sampled from the correlated inputs really matched that coefficient? @RISK risk simulation software enables you to see the actual simulated coefficients after a simulation – a very useful tool when auditing results. Just click on a cell in the correlation matrix after a simulation to see the actual simulated coefficients and their scatter plots. The coefficients you defined are also there for comparison so you can see how close it was.

Here is a short video demonstrating this:

» View short videos on recently added @RISK risk software features