"Also, these packages don't account for the possibility of a few years of consecutive bear market right after someone retires. How about the probability of having some years of bear market during retirement? I've no idea how the programs calculate the federal and state income taxes, how about capital gains taxes?"As a structural engineer, I need to know the assumptions before I trust any number. All of these fancy looking software packages are black boxes with limited user inputs, e.g., Boldin doesn't allow standard of deviation to be entered. This is a huge drawback in my opinion. Also, these packages don't account for the possibility of a few years of consecutive bear market right after someone retires. How about the probability of having some years of bear market during retirement? I've no idea how the programs calculate the federal and state income taxes, how about capital gains taxes? I can go on and on.
I've developed a spreadsheet (with no bells or whistles) that does Monte Carlo simulation. I use Portfolio Visualizer (or a program that I've written in open source R programming language) to calculate the mean & standard of deviation of my portfolios. (My R program also calculates the worst negative return for any portfolio that I specify with ticker symbols & weights.)
I haven't spent time yet to mess around with IRMMA, so I assume the amount for a higher marginal tax bracket, which I will not reach (even with RMD and Roth conversion). It calculates the taxes (federal and Ohio) according to the 2025 rates - I can also trig higher taxes if we revert to 2017 rates if TCJA is not extended. I can specify a number of years of bear market, or randomly select bear market years based on a preselected probability. I can also enter Roth conversion amounts in any year I select. For a selected life expectancy, it calculates various percentiles of the balances of tIRA, a brokerage account, and Roth IRA.
Yes, my spreadsheet assumes a normal distribution, which is a huge (and wrong) assumption. To the best of my knowledge, all the programs also assume a normal distribution for Monte Carlo simulation. I know I'm fooling myself by trying to be "precise" about many unknowns and the future. But at least I can control what assumption (right or wrong) I choose!
I'm happy to share the spreadsheet (which requires Excel for Monte Carlo simulation) with anyone who's interested.
Pralana can handle all of these.
"I haven't spent time yet to mess around with IRMMA, so I assume the amount for a higher marginal tax bracket, which I will not reach (even with RMD and Roth conversion)."
Pralana handles this.
"It calculates the taxes (federal and Ohio) according to the 2025 rates - I can also trig higher taxes if we revert to 2017 rates if TCJA is not "
This also.
"I can also enter Roth conversion amounts in any year I select. For a selected life expectancy, it calculates various percentiles of the balances of tIRA, a brokerage account, and Roth IRA."
It can also handle these as you desire and also has the ability to 'auto optimize' both Roth comversions and drawdown strategy as an optional 'toggle' for review. And a host of other things not mentioned so far.
Statistics: Posted by smitcat — Sun Jan 19, 2025 5:35 pm — Replies 30 — Views 3853