Importance of the sequence of returns for calculating SWR

Financial Independence/Retire Early -- Learn How!
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

I pretty much agree with you on most of this. Let's just say if we get into a gray area on board rules or ethics I'll do what I've always done....

I don't mean for anyone to get the idea that I was concerned about ElSupremo's board policies in any way. I wasn't trying to say that.

My concern was over where we draw the line on what is a personal comment and what is legitimate board business. I personally endorse the idea of not permitting any unnecessary "digs" at any posters, whether they post here or elsewhere. I think it's a big plus that ES is willing to step in here and help us in the event that this sort of thing ever gets to be a problem.

I just wasn't exactly clear on what Dagrims was trying to communicate. It may be that he (or she) was saying exactly what I am saying, that purely personal comments should not be permitted. My concern stems from an impression I have that some at the TMF board viewed my request to have intercst's posting privileges revoked as some sort of "personal" attack on him. I do not view it that way at all.

My reason for putting up that post is that he has wasted six months of the board's time with his posting tactics, and I believe that at some point that board must move on with him or without him. My preference is with him. But I do not have it in my power to force him to comply with the posting rules that apply over there.

To me, the "deception" and the "con" and the "fraud" are not personal matters. They are interfering with legitimate board business. I have no intent of coming over here and complaining about them as any sort of regular practice. It would be "disruption" on my part to do something like that. But I am going to make a pitch for people who do good work here to come over to the TMF board and help out when we proceed to Phase Two on Safe Withdrawal Rates. I may be required as part of that pitch to describe some of the deception that took place and why it is a big deal for people trying to learn how to put together a realistic Retire Early plan.

That's why I felt a need to respond to the Dagrims statement. I didn't want anyone to be surprised when I come forward and make this case at some later point. It's unfortunately become part of the project of telling the truth about withdrawal rates at this point. The whole point of my proposal from last Monday, that the REHP board appoint two referees to deal with this sort of situation, is that it would take this deception issue off the table and allow that board to return to discussion of issues directly relevant to early retirement. That's where I am trying to push things.

I would not say that the proposal I put forward exactly caught fire, however. JWR1945 did his part, and BenSolar did his, and StubbleJumper did his, and Galagan did his. But there is no board consensus at this time for Plan A. So for now, we are stuck on Plan B. I'm not trying to tell ElSupremo how to run his board, and I have no complaints re anything he had done or said. I just want Dagrims and the large number who I believe he speaks for to see why there is a need for responsible board members over there to come forward and ask that the right thing be done. The sooner that responsible parties over there do that, the better it will be for everyone involved, including intercst, in my opinion.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

Should there be any doubt, I am on record and still assert that all of hocus's comments in this manner are accurate and true. Any objective review of the record...although time consuming...will verify this.

Have fun.

John R.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

Oops! ...in this manner... should be ...in this matter...

Have fun.

John R.
rkmacdonald
* Rookie
Posts: 1
Joined: Sat Dec 07, 2002 7:33 am
Location: Plano, TX (near Dallas)

Re: Importance of the sequence of returns for calculating SW

Post by rkmacdonald »

raddr wrote: Wanderer and I were discussing awhile back how sensitive the SWR ("safe" withdrawal rate) is to the sequence of returns from year to year. I played around with Shiller's market data since 1871 and found that merely switching two years in the sequence could make a huge difference.


I think that what your really getting into here is the essence of Monte Carlo analysis. If you go to one of the websites that uses this technique, you will see that the max SWR is under 3%. At the T .Rowe Price site, the result is 2.64% for a 99% survival rate for 40 years ( http://www3.troweprice.com/ric/RIC/ ). This would be the absolute worst case based on historical data. It still doesn't mean that it couldn't be worse in the future, but I have a lot more faith in this approach than the one used on the REHP.

Also, check out this article: http://moneycentral.msn.com/articles/re ... s/8417.asp
RK
raddr
*** Veteran
Posts: 265
Joined: Mon Nov 25, 2002 3:25 am
Contact:

Post by raddr »

Hi RK, welcome to the board! :D

I think that what your really getting into here is the essence of Monte Carlo analysis. If you go to one of the websites that uses this technique, you will see that the max SWR is under 3%.

Yes but there are problems with Monte Carlo analysis, namely lack of mean reversion. I've looked at this and you might be interested in some work I've done on it:

http://groups.msn.com/IndexFunds/genera ... 7245748001

http://groups.msn.com/IndexFunds/firebo ... 9068563610

As you can see from the second post the inputs are very sensitive to expected returns and volatility.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

At the T .Rowe Price site, the result is 2.64% for a 99% survival rate for 40 years ( http://www3.troweprice.com/ric/RIC/ ). This would be the absolute worst case based on historical data.

One of the criticisms that has been raised re the intercst SWR study (but not by me) is that it is produced through DataMining, and therefore the results are not reliable. Intercst acknowledges that the study is an exercise in datamining, but defends his use of this statistical technique in this particular case on grounds that what he is doing is really an example of reverse datamining.

The claim is that he is not looking for a particularly good result (as was the case with the Foolish Four strategy that DataSnooper exposed at the TMF site) but a particularly bad result, the worst case scenario of all those that turned up in the 130 years of historical data examined. I personally believe that this defense is a valid one, so long as intercst makes every effort possible to identify the true worst possible case.

But this is precisely what her failed to do. There are all sorts of worst case scenarios that intercst failed to take into account in his study. One is the possibility that someday stocks might reach a level of overvaluation higher than they ever had in the 130 years he examined. That in fact happened in the late 1990s, and the failure to take this reality into account means that the study does not in fact consider the worst case scenario. Bernstein's analysis does factor in the overvaluation levels of the late 1990s, and that is why he came up with a SWR of 2 percent.

Another worst-case scenario not examined is the one pointed to by Raddr at the open of this thread--the fact that returns in the future may be the same as in the past, but the process by which those returns are achieved may not play out in the precise same way. Stocks may provide an average return of x, but Year Six may get switched with Year Five. Those little switches can cause a big drop in the safe withdrawal rate, as raddr shows with data. What he is describing is not a future worse than what we have seen in the past. The only investors who will perceive it as being "worse" are those who rely on the intercst study in planning their retirements.

These are not the only worst case scenarios that intercst failed to account for, but they are two of the most important ones.

Raddr expressed some surprise at the fact that changing the sequence of returns a small amount causes only slight changes upward in the SWR in some cases but relatively large changes downward in others. I believe that the cause of this disparity is intercst's use of the datamining technique. This is the exact criticism that is usually made of the technique, that it produces results that appear on the surface to be statistically valid but which are biased because of the desire on the part of the "researcher" to come to one particular conclusion rather than to just permit the data to speak for itself.

What makes it confusing is that in this case the usual signs of datamining are being shown to us in a mirror. Everything is reversed because the goal of the exercise was not to come up with the best possible answer (a high stock return), but the worst possible answer (the worst case scenario). But I believe that the bias with which the exercise was conducted produces the same sort of flaws as the more common form of datamining. Here the bias is not that intercst looked for good results that are not statistically likely to repeat (as in the Foolish Four strategy), but that he ignored bad results that would have pulled the safe withdrawal rate downward.

Does anyone see what I am getting at in saying that the root problem with the intercst study is its reliance on datamining (and a failure to do what would be required to make datamining appropriate in this case)? Am I right that the reason for the large drops in SWR from small sequence changes pointed out by Raddr is the use of a form of "datamining" in determining the SWR? I am not a statistics expert, or anything remotely close to it. So I do not want to make this claim on the other board if there are not others more knowledgable on this stuff agreeing with the essence of what I am saying here.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

At the T .Rowe Price site, the result is 2.64% for a 99% survival rate for 40 years ( http://www3.troweprice.com/ric/RIC/ ). This would be the absolute worst case based on historical data.

One of the criticisms that has been raised re the intercst SWR study (but not by me) is that it is produced through DataMining, and therefore the results are not reliable. Intercst acknowledges that the study is an exercise in datamining, but defends his use of this statistical technique in this particular case on grounds that what he is doing is really an example of reverse datamining.

The claim is that he is not looking for a particularly good result (as was the case with the Foolish Four strategy that DataSnooper exposed at the TMF site) but a particularly bad result, the worst case scenario of all those that turned up in the 130 years of historical data examined. I personally believe that this defense is a valid one, so long as intercst makes every effort possible to identify the true worst possible case.

But this is precisely what her failed to do. There are all sorts of worst case scenarios that intercst failed to take into account in his study. One is the possibility that someday stocks might reach a level of overvaluation higher than they ever had in the 130 years he examined. That in fact happened in the late 1990s, and the failure to take this reality into account means that the study does not in fact consider the worst case scenario. Bernstein's analysis does factor in the overvaluation levels of the late 1990s, and that is why he came up with a SWR of 2 percent.

Another worst-case scenario not examined is the one pointed to by Raddr at the open of this thread--the fact that returns in the future may be the same as in the past, but the process by which those returns are achieved may not play out in the precise same way. Stocks may provide an average return of x, but Year Six may get switched with Year Five. Those little switches can cause a big drop in the safe withdrawal rate, as raddr shows with data. What he is describing is not a future worse than what we have seen in the past. The only investors who will perceive it as being "worse" are those who rely on the intercst study in planning their retirements.

These are not the only worst case scenarios that intercst failed to account for, but they are two of the most important ones.

Raddr expressed some surprise at the fact that changing the sequence of returns a small amount causes only slight changes upward in the SWR in some cases but relatively large changes downward in others. I believe that the cause of this disparity is intercst's use of the datamining technique. This is the exact criticism that is usually made of the technique, that it produces results that appear on the surface to be statistically valid but which are biased because of the desire on the part of the "researcher" to come to one particular conclusion rather than to just permit the data to speak for itself.

What makes it confusing is that in this case the usual signs of datamining are being shown to us in a mirror. Everything is reversed because the goal of the exercise was not to come up with the best possible answer (a high stock return), but the worst possible answer (the worst case scenario). But I believe that the bias with which the exercise was conducted produces the same sort of flaws as the more common form of datamining. Here the bias is not that intercst looked for good results that are not statistically likely to repeat (as in the Foolish Four strategy), but that he ignored bad results that would have pulled the safe withdrawal rate downward.

Does anyone see what I am getting at in saying that the root problem with the intercst study is its reliance on datamining (and a failure to do what would be required to make datamining appropriate in this case)? Am I right that the reason for the large drops in SWR from small sequence changes pointed out by Raddr is the use of a form of "datamining" in determining the SWR? I am not a statistics expert, or anything remotely close to it. So I do not want to make this claim on the other board if there are not others more knowledgable on this stuff agreeing with the essence of what I am saying here.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

If anyone knows how to fix the columns that I somehow messed up with my last post, I'd appreciate it. I'd like to get some responses on the question I posed here, and I am concerned that the columns are now such a mess that no one is going to be able to read my long post.

Also, my post went up twice. That is the result of a message I received that the post had not transmitted successfully. If someone (ES?) has the power to delete one of the two identical posts, that would be a plus.
wanderer
*** Veteran
Posts: 363
Joined: Tue Nov 26, 2002 9:33 am
Location: anytown, usa

Post by wanderer »

you can delete it yourself, hocus, by clicking the "x" in the upper right hand corner.

formatting columns? i think most folks are transferring/linking jpg and gif images. MUCH better than tmf.

wanderer
regards,

wanderer

The field has eyes / the wood has ears / I will see / be silent and hear
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

Formatting columns? i think most folks are transferring/linking jpg and gif images. MUCH better than tmf.

I agree that there is some wonderful functionality at this place.

My concern, though, is that on my computer screen I need to scroll across the page to read the posts in this thread and some others. I'm worried that some might not want to go to the trouble.

Is this a problem with my particular computer screen? Perhaps it is just not big enough?

If it's a general problem, it would be nice if there were some way to have both the charts, which obviously add value, and yet also have the text posts (the only ones I am capable of offering!) appear in a way in which there is not a need to navigate horizontally.

If this is not a problem for others, please just don't worry about it. I am able to navigate horizontally if that is what is needed to permit the charts, and I see the value in the functionality that allows us to provide the charts.
User avatar
ElSupremo
Admin Board Member
Posts: 343
Joined: Thu Nov 21, 2002 12:53 pm
Location: Cincinnati, Ohio

Post by ElSupremo »

Greetings Hocus :)

Looks like all of your questions were answered. Thanks folks
"The best things in life are FREE!"

www.nofeeboards.com
User avatar
ataloss
**** Heavy Hitter
Posts: 559
Joined: Mon Nov 25, 2002 3:00 am

Post by ataloss »

I haven't been able to post tables because all the spacing has been removed (I am not sure if it a limitation of the board or my fault.)

I have posted links to wide jpgs that have resulted in a need to scroll. I don't know how th tell in advance if this will happen.

I have been amusing myself (at least one person is amused) by using odd place names for my location. When I used Llanfairpwllgwyngyllgogerychwyrndrobwll Llantysiliogogogoch, Wales the software widened the left column to accomodate this shifting all the other text to the right edge of the screen.

This board is great but not perfect (yet)
Have fun.

Ataloss
raddr
*** Veteran
Posts: 265
Joined: Mon Nov 25, 2002 3:25 am
Contact:

Post by raddr »

I have posted links to wide jpgs that have resulted in a need to scroll. I don't know how th tell in advance if this will happen.

Right click on the image, choose "Properties", and look at "Dimensions" for pixel size. Most monitors will display an 800X600 image fine. If it is much larger than that then I would resize the image before posting.
User avatar
ElSupremo
Admin Board Member
Posts: 343
Joined: Thu Nov 21, 2002 12:53 pm
Location: Cincinnati, Ohio

Post by ElSupremo »

Greetings ataloss :)
This board is great but not perfect (yet)


Getting a little greedy aren't we? :PI think using "Llanfairpwllgwyngyllgogerychwyrndrobwll Llantysiliogogogoch" is a little excessive.(Although I was amused. :lol:) Also raddr's suggestion of resizing the images seems a good one. Perhaps down to the point where everything is legible.
"The best things in life are FREE!"

www.nofeeboards.com
User avatar
ataloss
**** Heavy Hitter
Posts: 559
Joined: Mon Nov 25, 2002 3:00 am

Post by ataloss »

using an 800x600 I am think a width up to about 600 can be supported. This will work for images that we upload to anoter site but in terms of linking to other images you sort of get what you have. For some reason when I postred this 600 pixel wide image the text is wider than the image. :cry: Maybe the limit is more like 500. BTW I think these are baby boomer bikers :)
Have fun.

Ataloss
Post Reply