Re: Jaye Jarrett Safe Withdrawal Study
 04-30-2003, 06:20 AM #21 Recycles dryer sheets   Join Date: Nov 2002 Posts: 398 Re: Jaye Jarrett Safe Withdrawal Study Mikey has the right concept about how Monte Carlo simulation works. In the case of modelling returns on various classes of investments, it does the following: Say the user wants to model a ten year period, with 50% of assets in stocks and 50% in corporate bonds. For the first year, the program generates a number representing the total return on stocks for that year. This number is "semi-random" in the sense that the program is instructed to generate numbers that will, collectively, have a mean (average) equal to the historical average annual return on stocks, with a standard deviation equal to the historical standard deviation of the annual returns on stocks. The program then applies this value to the 50% of the "portfolio" that is in stocks, to yield a year-end value for the stock portion of the portfolio. In a similar fashion, the program generates a semi-random value for the return on bonds for year #1, and applies that to the 50% of the portfolio value that is invested in bonds. The "new" values of the stocks and the bonds are then added, and an annual withdrawal amount selected by the user is subtracted. Then, the portfolio is mathematically "rebalanced" so that there is again 50% in stocks and 50% in bonds at the start of year #2. This process is repeated for each of ten years, producing a particular value for the portfolio at the end of year #10. This same process is repeated numerous times (500 in the case of the TRP model), producing a large set of different "possible" values for the portfolio at the end of year #10. The presumed "chance of success" is determined by the number of results that are greater than zero. For example, if 450 results are greater than zero, there is considered to be a 90% chance that the investment strategy would be successful in the sense of supporting the input withdrawal rate over a ten year period. But sometimes the program will produce, say, 441 "successful" results and sometimes it will produce, say, 465 "successful" results. And in an extremely rare case, it could produce 500 "successful" results with the same input data ! From a standpoint of statistical purity, a problem with the relatively "simple" Monte Carlo model that I described is that it does not account for (a) correlation between returns on different asset classes (for example, the fact that annual stock and bond returns are somewhat correlated) nor (b) "serial correlation" within each asset class (the tendency for returns to vary cyclically from year to year). I suppose that an professional statistician could create a Monte Carlo simulation that would take these statistical interrelationships into account, but I doubt that the TRP model does. Interestingly, the FIRECalc model does account for these statistical interrelationships, to the extent that they are embodied in the historical data that FIRECalc uses. Conceptually, the major risk with FIRECalc is that the data that it uses might contain one or more extreme "outlier" years that would make its "worst case" results too pessimistic. But I don't think so, since I like to plan for what I consider to be a "reasonable worst case" scenario. __________________
 Join the #1 Early Retirement and Financial Independence Forum Today - It's Totally Free! Are you planning to be financially independent as early as possible so you can live life on your own terms? Discuss successful investing strategies, asset allocation models, tax strategies and other related topics in our online forum community. Our members range from young folks just starting their journey to financial independence, military retirees and even multimillionaires. No matter where you fit in you'll find that Early-Retirement.org is a great community to join. Best of all it's totally FREE! You are currently viewing our boards as a guest so you have limited access to our community. Please take the time to register and you will gain a lot of great new features including; the ability to participate in discussions, network with our members, see fewer ads, upload photographs, create a retirement blog, send private messages and so much, much more! Join Early-Retirment.org For Free - Click Here
Re: Jaye Jarrett Safe Withdrawal Study
 04-30-2003, 08:15 AM #22 Thinks s/he gets paid by the post   Join Date: Dec 2002 Posts: 3,875 Re: Jaye Jarrett Safe Withdrawal Study Re. "reasonable worst case scenario", when doing any sort of long term ER financial planning (almost a daily event), I always start with the worst case scenario and go back from there. Works for me. __________________
Re: Jaye Jarrett Safe Withdrawal Study
 04-30-2003, 03:19 PM #23 Give me a museum and I'll fill it. (Picasso)Give me a forum ...   Join Date: Apr 2003 Location: Hooverville Posts: 22,542 Re: Jaye Jarrett Safe Withdrawal Study Thank you Ted for your detailed observations and John Galt for your pithy comments. Ted is certainly correct that ignoring inter-correlations between asset classes is a very big flaw in any model. CAPM suggests that the risk free rate of interest that underlies valuations of both these classes guarantees at least some correlation over time. (Anything can happen in the short run!) Interestingly, correlation doesn't matter much, nor does variability, if there are no near term demands for withdrawals. It's the toll of taking money out as the portfolio value drops that kills. As a matter of personal history, I am in roughly my 20th year of early retirement. (Well, not so early anymore.) I began with modest resources, and raised two children, although I did not pay for their college educations. But all the usual middle class things were done-health insurance, braces, music lessons, etc. My resources were never depleted, but I admit there were some very good years in the last twenty! My technique paid a lot of attention to inter-correlations of issues and industries, but none to asset classes per se. I looked for individual securities, whether they are bonds, stocks, or partnership units, whatever, that I thought would prosper independently of the general market, and that would likely be on different cycles from one another. I wanted to have at least 20 or so of these. I also tried very hard to stick with competent and honest company managers. I made myself resist the urge to average down. I felt this rule was necessary to keep a mistake from sinking the ship. Some issues I have held for close to 20 years. During all this time, I took out quite a bit more than the safe withdrawal amount, sometimes using margin temporarily. In retrospect, I was safe, before the fact probably not. Now I have fewer obligations, and am moving toward a much lower withdrawal rate. Partly because it isn't so hairy. And partly because I am not as sure there are many bulletproof opportunities around. About 2 years ago I bought some TIPs, and some I-bonds. IMO, these were a golden opportunity for a truly passive but very safe investment. At current quotes, I don't find them particularly atractive. It is certainly possible that real interest rates will rise from here, and cause loss of principle in the TIPs, or poor returns in the I bonds. Likewise, I feel very negative about any fixed income security with more than a year or 2 to maturity. Some of this may be OT to safe withdrawal, in that it isn't mathematically based, but I wanted to offer it as one real world example of a different method that requires care, but can work. (Of course, this may be that outlier that the Big-Simulator-in-the-Sky occasionally throws out!) Mike __________________ "As a general rule, the more dangerous or inappropriate a conversation, the more interesting it is."-Scott Adams
Re: Jaye Jarrett Safe Withdrawal Study
 04-30-2003, 03:35 PM #24 Thinks s/he gets paid by the post   Join Date: Dec 2002 Posts: 3,875 Re: Jaye Jarrett Safe Withdrawal Study Hi Mike! Good stuff! I semiretired in 1993 and fully rerired in 1998. I am now 58. The only common stock I owned in the last 10 years was what I acquired as a result of some stock options from a company I did some short term work for in the late 90s. Since then I have owned preferred stock, bonds, CDs, Money market accounts, real estate, corporate notes, etc. I have yet to touch my "base". In fact (like Paul Terhorst) my net worth has grown since I retired. Am I smart or just lucky? Not sure, maybe a little of both.
Re: Jaye Jarrett Safe Withdrawal Study
 05-01-2003, 12:48 AM #25 Early-Retirement.org FounderDeveloper of FIRECalc   Join Date: Jun 2002 Posts: 1,826 Re: Jaye Jarrett Safe Withdrawal Study You can test your own assumptions about rate of return, standard deviation of the equities, inflation rate, and so forth, versus your own withdrawal plan, using a Monte Carlo simulator that tests 10,000 runs, and displays the answers in graphical form. See http://www.geocities.com/nandmmeyer/...drawalSim.html Dory36 __________________ Often uninformed, seldom undecided. Twenty years from now you will be more disappointed by the things you didn't do than by the ones you did do. So throw off the bowlines. Sail away from the safe harbor. Catch the trade winds in your sails. Explore. Dream. Discover. Mark Twain
Re: Jaye Jarrett Safe Withdrawal Study
05-01-2003, 06:46 AM   #26
Recycles dryer sheets

Join Date: Nov 2002
Posts: 398
Re: Jaye Jarrett Safe Withdrawal Study

The calculator referenced by Dory36, like the one by T. Rowe Price, does not provide the flexibility that FIRECalc does, in terms of varying the principal amount and withdrawal rate. It also does not account for the way that inflation has historically correlated with stock and bond returns to affect the annual real rate of return on these assets.

It does present an interesting idea as to a withdrawal strategy, but the "catch" is that in certain years of poor market performance, a person would need to curtail their withdrawal amount substantially. This largely defeats the whole point of the retirement planning exercise, which is to identify a fairly constant amount that may be "safely" withdrawn. A person could compensate for this by establishing a separate "rainy day" fund, but it too would need to be invested in some type of assets, and the calculator should ideally account for the performance of that too!

So I still think that FIRECalc is "best" if used with discretion, but these other calculators may provide some ideas as to how FIRECalc could be made even more flexible.

Quote:
 CAPM suggests that the risk free rate of interest that underlies valuations of both these classes guarantees at least some correlation over time.
As an adherent of monetarist economic theory, I would add that the common factor that determines the fluctuations in the valuations of different asset classes, and in the inflation rate, is the rate of growth of the money supply. The long-term real (inflation-adjusted)rates of return are determined primarily by the real growth in the output of the economy.
__________________

 Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)