Formula:
where s is the standard deviation of the forecast, k is the kurtosis, n is the number of trials, and z is the z value based on the specified confidence level (from Run Preferences, select Trials).
