How Do I Replicate this Asset Allocation in FIRECalc?

[nico08]

Hello. My current asset allocation is roughly as follows-

Large Cap Stock 34%
Mid/Small Cap Stock 9%
International Stock 18%

Investment Grade Bond 21%
High Yield Bond 4%
International Bond 7%

Short-term Securities 7%

I want to replicate this asset allocation in FIRECalc.
In the “Your Portfolio” tab in FIRECalc, I look at the “mixed portfolio” radio button. The asset allocation choices are- US Micro Cap, US Small, US Small Value, S&P 500, US Large Value, US LT Treasury, LT Corporate Bond, 1 Month Treasury.

What percentages do you think I should use for the available FIRECalc asset allocation portfolio choices, based on my actual current portfolio asset allocation identified above? I understand there wont be a perfect match of my current assets to the asset classes available in FIRECalc, but I would like to know your guestimate.

As a side note, it looks like my success rate is higher when I use the “mixed portfolio” choice compared to using the “Total market” choice using the Long Interest Rate and 60 percent in equities. Is this because in FIRECalc I use 1871 as data start date in the “Total market” section compared to using the default 1927 data start date in the “mixed portfolio” section of FIRECalc?

Thank you for your insight.

 

[pb4uski]

I have similar issues in that I have substantial international equities and bonds (and similar proportions). I think the best you can do is either go 60/40 equities/long interest rate in the total market box or go 61/32/7 between S&P 500/LT Corporate Bond/1 month Treasury in the mixed portfolio box.

I have always found the option in the mixed portfolio strange... why have US Large Value but not US Large Growth? US Small and US Small Value but not US Small Growth?

 

[Huston55]

I have similar issues in that I have substantial international equities and bonds (and similar proportions). I think the best you can do is either go 60/40 equities/long interest rate in the total market box or go 61/32/7 between S&P 500/LT Corporate Bond/1 month Treasury in the mixed portfolio box.

OP-Do you think it really makes any difference? Given the output curves & success %, I'd say it's likely within the margin of error; remember, it's a prediction.

 

[pb4uski]

Actually it makes a significant difference.

Total market with 61% equities and remainder long interest rate... all other assumptions default... 95.7% success rate.

Same as above except with mixed portfolio and 61/32/7 S&P 500/LT Corporate Bond/ 1 month Treasury... 83.7% success rate.

I was quite surprised by the magnitude of the difference.

 

[Huston55]

Actually it makes a significant difference.

Total market with 61% equities and remainder long interest rate... all other assumptions default... 95.7% success rate.

Same as above except with mixed portfolio and 61/32/7 S&P 500/LT Corporate Bond/ 1 month Treasury... 83.7% success rate.

I was quite surprised by the magnitude of the difference.

Wow!

Color me surprised.

However, I am in the (Bernstein?) school of, 'the last 10%+/- is irrelevant because of real life things completely out of the ordinary. (Plague, WW3, zombie apocalypse, zombie president apocalypse, etc.)

 

[nico08]

Hi Pb4uski. Thank you for your recommendations, that is what I was looking for. Until recently, I used the 60/40 equities/long interest rate in the total market box. And when I used a version of the "mixed portfolio" that was somewhat similar to the 60/40 mix, it produced a different (better) percent likelihood of portfolio survival.

I have similar confusion about the "mixed portfolio" asset choices. The choices don't seem to cover the full range of asset classes. Like you said, there may be a class for US large growth, but not US large value and a choice for any international investment assets (which is a part of a diversified portfolio) is missing.

You validated the confusion that I have with that part of FIRECalc. Thanks.

 

[bmcgonig]

Actually it makes a significant difference.

Total market with 61% equities and remainder long interest rate... all other assumptions default... 95.7% success rate.

Same as above except with mixed portfolio and 61/32/7 S&P 500/LT Corporate Bond/ 1 month Treasury... 83.7% success rate.

I was quite surprised by the magnitude of the difference.

The long bond has been in a bull for over 30 yrs. I've always assumed that to be the reason for Firecalc to do that. I never use long interest rates in Firecalc, does anyone else?


Sent from my iPad using Early Retirement Forum

 

[REWahoo]

The long bond has been in a bull for over 30 yrs. I've always assumed that to be the reason for Firecalc to do that. I never use long interest rates in Firecalc, does anyone else?

Try running FIRECalc using the the four options ( Commercial Paper, Long Interest Rate, 30 Year Treasury, or 5 Year Treasury) and compare the results. Please let us know if you see a significant difference.

 

[bmcgonig]

Try running FIRECalc using the the four options ( Commercial Paper, Long Interest Rate, 30 Year Treasury, or 5 Year Treasury) and compare the results. Please let us know if you see a significant difference.

I assume I will. Especially between 30 and 5 . My point was who around here puts their bond money in long dated instruments.

Sent from my Nexus 6P using Early Retirement Forum mobile app

 

[pb4uski]

It depends on what you mean by long. 30 years... very few.... 10 years... a lot.

 

[MBSC]

Actually it makes a significant difference.

Total market with 61% equities and remainder long interest rate... all other assumptions default... 95.7% success rate.

Same as above except with mixed portfolio and 61/32/7 S&P 500/LT Corporate Bond/ 1 month Treasury... 83.7% success rate.

I was quite surprised by the magnitude of the difference.

Until recently, I used the 60/40 equities/long interest rate in the total market box. And when I used a version of the "mixed portfolio" that was somewhat similar to the 60/40 mix, it produced a different (better) percent likelihood of portfolio survival.

FWIW, total market has more data points. Mixed portfolio is set to 1927 and runs 49 cycles. Total market defaults to 1871 and runs 116 cycles. Changing total market to 1927 runs 60 cycles with a 91.7% success rate.

 

[haha]

Actually it makes a significant difference.

Total market with 61% equities and remainder long interest rate... all other assumptions default... 95.7% success rate.

Same as above except with mixed portfolio and 61/32/7 S&P 500/LT Corporate Bond/ 1 month Treasury... 83.7% success rate.

I was quite surprised by the magnitude of the difference.

Couldn't this be an artifact of a lot of very random things in the database? I would fall over if this were a stable difference.