Early Retirement & Financial Independence Community (http://www.early-retirement.org/forums/)
-   FIRECalc support (http://www.early-retirement.org/forums/f36/)
-   -   Question on interpreting FireCalc (http://www.early-retirement.org/forums/f36/question-on-interpreting-firecalc-68112.html)

 modhatter 08-26-2013 01:31 AM

Question on interpreting FireCalc

If I plug in withdrawal amounts and portfolio \$ and then go to page where you either let it do it's own thing, or drop down to where you would plug in a closer approx. of fund types and percentages, it comes out fine.

But when I go down to the last input on the Portfolio page and put in a % of a mean expected return, starting with 7% it fails miserably. I then have to bump that number up to 9% average return before I get a 100% success.

In putting in the 7%, I am thinking that I am saying that after x amount of years when all gains and losses are added, I averaged a 7% total return for each year. Is this correct?

When you calculate, are you inputting the expected future return expectations, or are you letting FireCalc put in the % Return based on past performance?

 ERD50 08-28-2013 07:21 AM

I tried that and if I take the std dev to zero, the success % goes way up. What does that tell us?

I have zero interest in MonteCarlo - who knows what data is being used? Is there any correlation between the different investments, inflation, etc?

Best to sick with historical, IMO.

-ERD50

Std dev is standard deviation. It is not a yield, but a measure of difference from the average mean (the bell shaped curve). It is a volatility measure.

Taking Std Dev to 0 means your portfolio matches the volatility of the market and over time your results will look like the bell curve. A small Std Dev means less volatility, a large one more volatility. Small = blue chip stocks, large = small company stocks.

-- Rita

 ERD50 08-28-2013 08:34 AM

Quote:
 Originally Posted by Gotadimple (Post 1353218) Std dev is standard deviation. It is not a yield, but a measure of difference from the average mean (the bell shaped curve). It is a volatility measure. Taking Std Dev to 0 means your portfolio matches the volatility of the market and over time your results will look like the bell curve. A small Std Dev means less volatility, a large one more volatility. Small = blue chip stocks, large = small company stocks. -- Rita
I know what STD DEV is - what I meant to ask was "What does it tell us about our portfolios if we enter various STD DEV values into this Monte Carlo calculator?"

It just seems like such a phony way to look at a portfolio - IMO, history tells us more, and of course its value is limited.

-ERD50

ERD,
I know YOU know, but in reading the OP, it doesn't appear that user knows what the field is for

"In putting in the 7%, I am thinking that I am saying that after x amount of years when all gains and losses are added, I averaged a 7% total return for each year. Is this correct?"

 ERD50 08-28-2013 09:27 AM

Well, it should still average 7%, regardless of the std dev. Hmmm, does Monte Carlo 'know' that a 20% loss must be followed by a 25% gain to get back to par?

This is why I find Monte Carlo useless. If I'm curious about how some pattern affects me, I will do my own spreadsheet where I can see everything. I have no idea what the Monte Carlo assumptions are. So it's hard to say why a certain number would provide a certain result. If you take STD DEV to zero, the graph should show the straight-line results, but a calculator can do that easily.

-ERD50

 modhatter 08-28-2013 12:19 PM

Last box to use in Firecalc Portfolio page says the following:(and you get to change three numbers - the total return % and the standard deviation % and the assumed inflation rate)

A portfolio with random performance, with a mean total portfolio return of ____% and variability (standard deviation) of ____%. Assume an inflation rate of ____%.

This is what I was referring to. I understand what standard deviation is. I was referring to the automatic 10% "total portfolio return" they plug in. I know that the average return over a long period of time in the past averaged around 10%. However all the financial pundents and economist projections of future growth for the US are not so optomistic.

Most financial planners seem to feel we will be more in the neighborhood of 7% for total portfolio return at least in the next decade. It was this number that I was adjusting down to 7%. I also played with the standard deviation, just to see, but that wasn't my question so much.

Just wondered what most were doing when they look at FireCalc in evaluating their portfolio. What are they plugging in. Since this is the only box that allows you to change the expected future total return.

Another factor effecting outcome would be very low bond yields. I am not sure about the past, but if we never had long periods of super low bond yields (coupled with lower dividends) then this could effect numbers as well, as there would be less income to draw from now than in the past, and more selling of stocks for income.

I am not sure of the bond "yield" history of the past, so I may be wrong on that one. I think the yield matters more when you are in the draw down phase (retirement) as far as it effects sustainability of the portfolio. Maybe it doesn't matter, though since when yields decrease, value increases, so bonds can be sold for the income. I think I'm confusing myself now.

 ERD50 08-28-2013 12:41 PM

Quote:
 Originally Posted by modhatter (Post 1353272) ... Most financial planners seem to feel we will be more in the neighborhood of 7% for total portfolio return at least in the next decade. .... Just wondered what most were doing when they look at FireCalc in evaluating their portfolio. What are they plugging in. Since this is the only box that allows you to change the expected future total return.
Even if you think you can predict market returns and inflation averages, can you predict their interactions and sequence?

I don't, so I don't 'plug in' anything. I look at what the worst of history has dished out (which has nothing to do with averages), and aim for a WR that would survive that historic worst case.

In terms of using FIRECalc that way, it means nothing if average future returns are lower than average past returns. What would be an issue is whether the future is worse than the worst case of the past. Obviously, that could happen, but it has nothing to do with averages. And I try to put some buffer in beyond the historic worst case.

BTW - here are the historically safe withdraw rates in FIRECALC - these all leave ~ \$20K in the ending portfolio:

Code:

```Years in Plan:    HSWR        5          13.00% 10          6.80% 15          5.18% 20          4.35% 25          3.84% 30          3.57% 35          3.44% 40          3.32% 45          3.23% 46          3.36%```
I stop at 46 years, as the % rises again due to dropping off bad starting years (1966).

-ERD50

 All times are GMT -6. The time now is 04:21 PM.