For me, unchecking "Assume correlated asset classes?" greatly increased my projected Monte Carlo outcome making it much closer than Historical, but still realistically lower. It also still performs below the three Bear Markets that are included. In reading the documentation on correlated asset classes, it seems like the logical better choice, but the outcomes for me are so dramatically below Historical and other calculators, I'll avoid using it in the future.
Greg,
It sort of sounds like you're not believing that the MC results are accurate, so I decided to play with this a bit myself. I just ran a test using the canned scenario 1 in the deliverable version of PRC2021, with these exceptions: I changed the assets to purely stocks and bonds with a 50/50 allocation in all accounts and changed the RROR for stocks and bonds to be the average of the historical real returns for stocks and bonds for the period from 1928-2020. These average ROR values are then used as the mean ROR in determining the random returns in the MC analysis. Then I ran both a MC and a historical analysis. In this scenario, the MC results are substantially better than historical results. I don't have the time to thoroughly analyze this finding at the moment but plan to study this in depth and include some discussion on this topic in the user manual for PRC2022.
How do your average ROR's as entered on the Financial Assets > Asset Classes page compare to the historic averages?
Thanks,
Stuart
Stuart,
Actually, I'm very much believing the MC results when I uncheck "Assume Correlated Asset Classes?". Checking that box gives me a far more pessimistic result dramatically below the Bear Market models you include, so reluctant to believe those results.
I've attached a screenshot of my Asset Class ROR and I'm planning for a 40 year retirement and a 60/40 equity/bond allocation. The asset allocation table produces an average of 3.14% ROR for me. I have inflation at 2.5%.
-Greg
