I noticed that while the UI says geometric mean, the backend actually treats any value which is input here as arithmetic mean. converts it to a geoemetric mean and then applies the provided geometric standard deviation when running Monte Carlo Simulations. In my opinion this is highly problematic since while the resulting distributions exhibit similar PDFs, the whole distribution is shifted. I only noticed this when entering global parameter uncertainty but I assume its a global issue.
Am I misunderstanding how this value should be used or is is actually an issue?
