mathjak107
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
- Joined
- Jul 27, 2005
- Messages
- 6,210
yes and no.
each time frame is looked in seperate context of a 30 year period in the trinity study , bill bengan's work and firecalc.
the results of each 30 year period are looked at . each period contains the best and worst of only its time frame.
then the worst period is pulled out and analyzed to see what could be drawn off it.
i believe 4% did fail in 1965-1966 depending who's data you use and there may have been one or 2 more so when all time frames are considered a percentage of pass fail can be arrived at.
hypothetically if you looked at 10 periods of time and one failed then you had a 90% success rate. the best of each period is not mixed into other periods of time when using historical data like firecalc. if we then take that failed period and see what draw rate let it pass you have the basis for determing a safe withdrawal rate.
we had 17 great years from 1987 to 2003 averaging almost 14% cagr for 17 years . but unless your time frame was part of that those number don't help your period .
in monre carlo simulatons you could pair the best of the gains with the worst of the gains or the worst gains with worst sequencingA and get combo's historical never saw,
i think the best example is the fact that while 1987 to 2003 were fabulous times for gains the time frame leading in were the worst of times.
20 years of poor markets followed by double digit inflation crushed retiiree's as well as few in the accumulation stage were able to save any money for when the best of times came.
so in that period the fact we had the greatest markets wasn't so great since few had much money invested because the time frame before was so poor.
the flaw in historical data is that may never play out in that order again.
each time frame is looked in seperate context of a 30 year period in the trinity study , bill bengan's work and firecalc.
the results of each 30 year period are looked at . each period contains the best and worst of only its time frame.
then the worst period is pulled out and analyzed to see what could be drawn off it.
i believe 4% did fail in 1965-1966 depending who's data you use and there may have been one or 2 more so when all time frames are considered a percentage of pass fail can be arrived at.
hypothetically if you looked at 10 periods of time and one failed then you had a 90% success rate. the best of each period is not mixed into other periods of time when using historical data like firecalc. if we then take that failed period and see what draw rate let it pass you have the basis for determing a safe withdrawal rate.
we had 17 great years from 1987 to 2003 averaging almost 14% cagr for 17 years . but unless your time frame was part of that those number don't help your period .
in monre carlo simulatons you could pair the best of the gains with the worst of the gains or the worst gains with worst sequencingA and get combo's historical never saw,
i think the best example is the fact that while 1987 to 2003 were fabulous times for gains the time frame leading in were the worst of times.
20 years of poor markets followed by double digit inflation crushed retiiree's as well as few in the accumulation stage were able to save any money for when the best of times came.
so in that period the fact we had the greatest markets wasn't so great since few had much money invested because the time frame before was so poor.
the flaw in historical data is that may never play out in that order again.
Last edited: