Is it legit to re-run when market rises?

[donheff]

Good grief what a tiring thread. The most useful thing I got out of it was a re-look at the Bernstein calculator from hell piece. My impression from way back was that he had made some conservative prediction that reduced returns over the coming years would lead to worse results than are present in the historical record. But it is clear looking at the article that he is not. Basically he is saying that he thinks we are heading for a lousy 30 year run and uses 1966 as his benchmark. I.E., from a Firecalx perspective he is saying we should expect to be on some of the lousy scenarios. That still leaves us in the 4% SWR range for 100% FC success. It seems to me that most of the Firecalc proponents around here tend to start there and back off to lesser SWRs of 3-3.5% (or switch to variable withdrawal approaches) to accommodate the possibility that the future will be even worse than the worst of the past.

 

[REWahoo]

It seems to me that most of the Firecalc proponents around here tend to start there and back off to lesser SWRs of 3-3.5% (or switch to variable withdrawal approaches) to accommodate the possibility that the future will be even worse than the worst of the past.

+1

While the FIRECalc antagonists tend to start and perpetuate threads foreseeing the falling of our financial sky and proposing...?

 

[Midpack]

It seems to me that most of the Firecalc proponents around here tend to start there and back off to lesser SWRs of 3-3.5% (or switch to variable withdrawal approaches) to accommodate the possibility that the future will be even worse than the worst of the past.

+1. While the FIRECalc antagonists tend to start and perpetuate threads foreseeing the falling of our financial sky and proposing...?

+2. But I don't want to be tiresome, I've already (over) explained my POV...

 

[bingybear]

It can be replicated - X number of people with the same inputs, will get the same results. That's a 'known science'.

so does this mean any code that can take the same inputs in 'known science'? time independent code yes.
I think we would agree that firecalc is a useful tool that back tests the inputs based on historical data. As an engineer we would look at expected variations and worst case based on physical laws... not just historical tests. Honestly I don't know how financial science is defined... so the term for firecalc may be right... but because a code gives the same output with the same input has less to to with financial science than having a time independent code. And this does not require it to be correct in the first place, just repeatable.

It is kind of the difference of being precise and being accurate.

 

[audreyh1]

Why make a "withdrawal" if it then just gets parked in a different place?

A critically important point:

The funds in your retirement portfolio are subject to market volatility. Supposedly there is sufficient equity exposure in that portfolio to offset long term inflation.

The "place" short term funds are parked in should be protected from that same market volatility. That's in fact the whole point!

 

[Bestwifeever]

A critically important point:

The funds in your retirement portfolio are subject to market volatility. Supposedly there is sufficient equity exposure in that portfolio to offset long term inflation.

The "place" short term funds are parked in should be protected from that same market volatility. That's in fact the whole point!

But we have 10 percent of our retirement portfolio already in a money market fund protected from that same market volatility. So why withdraw and repark it?

ETA we also would have to pay tax on the withdrawal which might not be true for you.

 

[audreyh1]

But we have 10 percent of our retirement portfolio already in a money market fund protected from that same market volatility. So why withdraw and repark it?

ETA we also would have to pay tax on the withdrawal which might not be true for you.

What is that money for? Is to rebalance the portfolio or to fund withdrawals? I think it's the former. You rebalance when you withdraw, right?

I prefer to be very clear about the purpose and time frame of each of my investments. Different time frames and investment purposes are managed separately. I'm not going to mix it up by having cash set aside for another purpose mixed into my AA calculation when I'm rebalancing my portfolio. I also don't include money set aside for some other purpose when I calculate my withdrawal.

No, I don't pay taxes for withdrawing cash.

Money that I don't spend - if I put in back in my retirement fund, I would consider that subject to the same 55/40/5 equities/bonds/cash AA volatility as the rest of my portfolio. Otherwise I would have to recalculate my AA with a larger cash allocation so I don't accidentally rebalance and dip into it too much. I simply choose not to do it that way.

I can see limiting withdrawals from an IRA to spending as it means less taxes (now). In that case I would track different "kitties" using a spreadsheet.

 

[ERD50]

Originally Posted by ERD50 View Post
It can be replicated - X number of people with the same inputs, will get the same results. That's a 'known science'.

so does this mean any code that can take the same inputs in 'known science'? time independent code yes. ...

Mathematics is a science. Computer programming is a science. That is what the historical reporters are.

I think we would agree that firecalc is a useful tool that back tests the inputs based on historical data.

Yes - and that is all it is! It even says so in the description!

As an engineer we would look at expected variations and worst case based on physical laws... not just historical tests. Honestly I don't know how financial science is defined... so the term for firecalc may be right... but because a code gives the same output with the same input has less to to with financial science than having a time independent code.

And here again, I think you are assigning something outside of its design criteria, and then claiming it doesn't work! How come my toothbrush won't backup my computer!

FIRECalc isn't designed to anticipate what would happen if the Feds raise interest rates for example, or any other calculation that an economist might ponder. It simply reports what has happened.

And this does not require it to be correct in the first place, just repeatable.

Except with public data, one can test for themselves if the output is correct. It is determinate.

It is kind of the difference of being precise and being accurate.

No, it really isn't.

-ERD50

 

[Bestwifeever]

What is that money for? Is to rebalance the portfolio or to fund withdrawals? I think it's the former. You rebalance when you withdraw, right?

I prefer to be very clear about the purpose and time frame of each of my investments. Different time frames and investment purposes are managed separately. I'm not going to mix it up by having cash set aside for another purpose mixed into my AA calculation when I'm rebalancing my portfolio. I also don't include money set aside for some other purpose when I calculate my withdrawal.

No, I don't pay taxes for withdrawing cash.

Money that I don't spend - if I put in back in my retirement fund, I would consider that subject to the same 55/40/5 equities/bonds/cash AA volatility as the rest of my portfolio. Otherwise I would have to recalculate my AA with a larger cash allocation so I don't accidentally rebalance and dip into it too much. I simply choose not to do it that way.

I can see limiting withdrawals from an IRA to spending as it means less taxes (now). In that case I would track different "kitties" using a spreadsheet.

I know DH and I are the least scientific and two of the most unsophisticated investors on these boards, so I realize we have different investment, withdrawal, and spending styles, Audrey. All is good

 

[harley]

And here again, I think you are assigning something outside of its design criteria, and then claiming it doesn't work! How come my toothbrush won't backup my computer!

I'm sure there's a relatively simple hack that would make that work, Philips Sonicare DiamondClean Toothbrush charges via rinsing glass or USB

 

[bingybear]

Mathematics is a science. Computer programming is a science. That is what the historical reporters are.

As an engineer we would look at expected variations and worst case based on physical laws... not just historical tests. Honestly I don't know how financial science is defined... so the term for firecalc may be right... but because a code gives the same output with the same input has less to to with financial science than having a time independent code.

And here again, I think you are assigning something outside of its design criteria, and then claiming it doesn't work! How come my toothbrush won't backup my computer!

My post was asking how firecalc is a science. you state because the program gives the same output for the same inputs for different people it is a known science'. My comment above was not to say that firecalc should do... but what I would expect of a science program.
I'm not trying to understand the term science from your argument. Because you get the same outputs from the same inputs makes firecalc a known science... would most computer programs not fit this definition?

The question is not what firecalc can do or should do... but what makes it a science? As I stated, I don't know how financial science is defined. I may just be missing that piece. Now if it because it is based on Computer science... that is being a program.... then is this forum a science?

 

[gcgang]

http://youtu.be/-FIMvSp01C8

SCIENCE!!!


Sent from my iPad using Early Retirement Forum

 

[ERD50]

...
The question is not what firecalc can do or should do... but what makes it a science? As I stated, I don't know how financial science is defined. I may just be missing that piece. Now if it because it is based on Computer science... that is being a program.... then is this forum a science?

I guess I'm just not following what distinction you are after, or why.

It seems clear what FIRECalc does - you provide inputs, it performs a series of computations on historical data and presents the data. It does it the same each time.

What does that have to do with any sort of 'financial science' and why does it matter?

But FWIW - from wiki:

Applied mathematics is a branch of mathematics that deals with mathematical methods that find use in science, engineering, business, computer science, and industry. Thus, "applied mathematics" is a mathematical science with specialized knowledge.

http://youtu.be/-FIMvSp01C8

SCIENCE!!!

YOU BLINDED ME!!!

Thanks, I needed that, hadn't heard it in a long time (well, a little goes a long way )

-ERD50