nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Planning Simulation Probability

brian_ecom

Quicken Classic - Windows Version (v R68.31)

So, I finally saw and tried out the Monte Carlo simulation on Quicken Classic for Windows. I recently had an engagement with a financial planner and wanted to compare results. My planner's software (not sure what he used) showed us a 99.9% success rate, while Quicken has us at 84% with the 10th percentile at $0. I've compared assumptions and do not see anything significant to warrant this difference (rate of return, inflation, expenses, taxes, life expectancy). In addition, I just plugged my financial data into an AI and get the same results as my planner.

I'm looking at my portfolio, it's a good amount, and I wondering how anyone can get above 95% using Quicken's model. My AI is saying that it's highly unlikely to go to $0 in my situation, much less a 16% chance.

Anyone else running this simulation and questioning the results?

Find more posts tagged with

Comments

Chris_QPW

There is a much longer thread that talks about this here:

https://community.quicken.com/discussion/7972995/lifetime-planner-monte-carlo-analysis-new

To directly answer your questions, yes, some people that have even got it to say 100%. But you have to understand that that percentage is probably the least important thing on the screen in my opinion. And off of the top of my head, I don't even know what it is calculating (maybe it was mentioned in the above thread). I took a quick look at it in the documentation and didn't see it.

You will notice that there are three graph lines going from optimistic to pessimist, so what is that "overall percentage"? You have in fact, three different numbers/results right there. Whereas I assume the planner/AI is using just one, maybe the middle ground one (and maybe that is what the main percentage Quicken is using, I'm not sure, I no longer track investments in Quicken so I can't go back and check it)

But one thing I will point out that I found and commented on early on in that thread is the "rate of return" you put into Quicken can change things drastically. And as I think of that, what you put in for inflation would do the same.

brian_ecom

Hey, thanks for responding. I did see that thread, and there was some discussion about what people thought about Quicken's results. I wanted to start a thread directly questioning and comparing Quicken's results. My withdrawal rate is 2.5% and Quicken has me going to $0, at least 10% of the time. In my AI, I asked what percentage of the time I went to $0 - out of 5,000 simulations I went to $0 exactly zero times.

I used the exact same rate of return and inflation (7% and 3% respectively). In fact, I made sure the AI matched Quicken planning assumptions. I know there is no guarantee on these simulations on any system, but what a radical difference.

If Quicken is correct, I really want to know since 10% of the simulations I run out of money. Do I radically change my lifestyle? Will I be shortchanging myself by a huge amount and miss out on a lot of life and gift giving if Quicken is as wrong and I start radically cost cutting.

I really appreciated, your feedback and insight on this topic.

Jim_Harman

A Monte Carlo analysis will take into account not only on the rate of return but also the standard deviation or spread of the expected returns. Think of it as a bell curve with the specified rate of return in the middle and the tails going out to the sides. If there is a wide spread of expected returns, there will be a higher percentage of outcomes that lead to either a lot of money left over or your assets going to zero.

The results you get will depend on all the underlying assumptions.

Chris_QPW

Speaking of assumptions, do either of these (Quicken and other tools) take into account things like RMD, which accounts are drawn from in what order? That kind of thing.

Because we definitely know Quicken's assumptions on these are currently wrong. For instance, the RMD date is set at 70 1/2, and for most people the account it takes anything it needs over that is the wrong type.

I would also want to say they call it Monte Carlo for a reason; it is like gambling in that you are getting the odds, and they are never 100% that you will succeed or fail for that matter. Just as I would only use Quicken's Lifetime Planner as a "ballpark estimate". It is certainly not going to be exact, especially some of the baked in wrong assumptions. And even if all of its assumptions were correct they idea that one will have the same spending over their retirement is almost never correct. What tends to happen is people spend more just after retirement and then as they age and can't do as much their spending goes down, and then maybe rises again at the end of life for medical reasons. But everyone's spending is of course going to be different.

Jim_Harman

… and many of the assumptions are un-knowable:

How long will my spouse and I live?
Will I need to live in a nursing home if so for how long?
How will investments perform over the next 30 or so years?
What will tax rates be in the future?
How much will Social Security be paying 20 years from now?

Chris_QPW

Yeah, and from what I see Monte Carlo simulation only varies the investment returns and then only using a "what has happened in the last X years" as boundaries.