Author Topic: Duration For FIRECalc  (Read 2904 times)

mr_orange

  • Walrus Stache
  • *******
  • Posts: 5612
  • Age: 42
  • Location: Round Rock, TX
Duration For FIRECalc
« on: July 20, 2016, 10:22:48 AM »
I know this is probably covered in other threads, but the search feature isn't working for me right now so I thought I would ask.  What time duration do you use in your FIRECalc estimates?  My probabilities are different for 30 years than they are for 50 so I wanted to see how other people thought about this.

I know that this isn't the end-all-be-all, but I figured I would go with the best general philosophy.  If anyone knows of other threads discussing this or other assumptions I should be making it would be helpful to link to them in this post. 

tonysemail

  • Pencil Stache
  • ****
  • Posts: 718
  • Location: San Jose, CA
Re: Duration For FIRECalc
« Reply #1 on: July 20, 2016, 01:12:46 PM »
this thread may be helpful
http://forum.mrmoneymustache.com/welcome-to-the-forum/50-60-year-retirement-timeframes-anyone-have-the-numbers/

I use 30 years because I'd rather have more sample points than longevity.
What question are you trying to answer in your simulation?
Just the success rate criteria or are you investigating some parameter like SWR?

Altons Bobs

  • Bristles
  • ***
  • Posts: 339
Re: Duration For FIRECalc
« Reply #2 on: July 20, 2016, 01:39:20 PM »
I was using 50 and then I found that 49 would give me worse results, so I started using 49 for worse case than 50. That's just me.

mr_orange

  • Walrus Stache
  • *******
  • Posts: 5612
  • Age: 42
  • Location: Round Rock, TX
Re: Duration For FIRECalc
« Reply #3 on: July 20, 2016, 01:40:40 PM »
Thanks.

I am mostly trying to get a sense for which to select given my age.  I'll probably err on the conservative side anyway so using a longer number is another good data point in addition to all the other data I am tracking.

mcneally

  • Stubble
  • **
  • Posts: 198
Re: Duration For FIRECalc
« Reply #4 on: July 20, 2016, 04:15:35 PM »
I was using 50 and then I found that 49 would give me worse results, so I started using 49 for worse case than 50. That's just me.

That's mathematically impossible as every scenario that fails within 49 years would also fail within the first 49 years of a 50 year period, plus however many additional scenarios fail in the last year of the 50 year period.

mr_orange

  • Walrus Stache
  • *******
  • Posts: 5612
  • Age: 42
  • Location: Round Rock, TX
Re: Duration For FIRECalc
« Reply #5 on: July 20, 2016, 05:15:09 PM »
For grins I ran my data today using 30-year and 50-year for my "lower-bound" and "upper-bound" rails, which are driven by a range in desired spending. 

Right now I have the following probabilities:

1.  30-year, lower-bound - 72.4%
2.  30-year, upper-bound - 2.6%
3.  50-year, lower-bound - 50.0%
4.  50-year, upper-bound - 0.0%

At the end of the year I am projecting to have:

5.  30-year, lower-bound - 99.1%
6.  30-year, upper-bound - 40.5%
7.  50-year, lower-bound - 94.8%
8.  50-year, upper-bound - 18.8%

I plan to work on my business full time after I "retire" from my W2 so the proper figures to use are a bit squishy.  My spending is, of course, a huge variable to nail down more.  For now I am comfortable working a few more years to increase spending because I have a pretty posh setup.  This may change though. 

Spork

  • Walrus Stache
  • *******
  • Posts: 5747
    • Spork In The Eye
Re: Duration For FIRECalc
« Reply #6 on: July 20, 2016, 05:19:50 PM »
I have automated runs against firecalc and cfiresim that run and graph daily.  I can't say I put a lot of thought into what number of years were statistically better or worse...  What I did was calculate the number of years to 2070.  That year, my wife will be 100 years old and I will most definitely be dust.  I always figured if we had good success rates for (2070 - current_year) years ... we were okay.

myhotrs

  • Stubble
  • **
  • Posts: 136
  • Age: 42
  • Location: Los Angeles
Re: Duration For FIRECalc
« Reply #7 on: July 20, 2016, 11:47:19 PM »
I'll add my numbers for analysis (and cuz its fun to run firesim)

FIRE today:
30 yrs: 91.8
50ys: 96.6

FIRE on my day next April:
30yrs: 95.9
50yrs: 96.6

Some of those seem strange, how the 50yr success rate doesn't change with all the $ I'll be adding in these last nine months.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 9563
  • Registered member
Re: Duration For FIRECalc
« Reply #8 on: July 21, 2016, 12:59:11 AM »
Use the worst of the two

Metric Mouse

  • Walrus Stache
  • *******
  • Posts: 5279
  • FU @ 22. F.I.R.E before 23
Re: Duration For FIRECalc
« Reply #9 on: July 21, 2016, 01:28:39 AM »
Use the worst of the two

And then pull the plug anyway. You'll have 50 years to sort it out.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28429
  • Age: -999
  • Location: Seattle, WA
Re: Duration For FIRECalc
« Reply #10 on: July 21, 2016, 02:02:32 AM »
Use the worst of the two

+1.  Longer has a sample size problem, shorter has a duration problem.
I am a former teacher who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and am now settled with three kids.
If you want to know more about me, this Business Insider profile tells the story pretty well.
I (rarely) blog at AdventuringAlong.com. Check out the Now page to see what I'm up to currently.