Just for the hell of it this is a portfolio longevity chart
- N = 1,000,000
- r = .04
- sd = .12
- spend rates are uniformly distributed between 2% and 12%
- horizon set to 100 years max where 100 is sorta infinity
Since this is one of the "shapes" of retirement that I set out to see, I wanted to see what it looked like with a really big N for the sim iterations. Looks like the smaller N of course, just more filled in. Also it's easier to see that 2% spend is almost, but not certainly, a perpetuity when returns are volatile. Interesting to ponder long retirements like early retirees in FIRE movement or long duration trusts or endowments. Didn't take too long to run, just a couple minutes or so.
Hi William, am trying to read your charts. does that mean those grew areas are also areas where failure happens?
ReplyDeleteReading horizontally for 1 spend rate, in cross section it would be a frequency or density for the likelihood that the portfolio has ended plus a distribution for those that lasted forever (at 100). So the grey dots would be a lower liklihood zone and the blue higher and red orange and yellow higher still. At 12pct spend there would be a distribution peaking around 5 years with a really long right tail of residual low level probabilities.
DeleteThanks for the explanation. Certainly a bit hard for me to digest but will try. Have a Merry Xmas to your family and yourself.
Delete