NoFeeBoards.com!

No Fees! No Ads! No Spam!

Log in Register FAQ Memberlist Search NoFeeBoards.com! Forum Index

Frequent Responses to Hocus- FRH-1- Invalidity
Goto page 1, 2  Next
 
Post new topic   Reply to topic    NoFeeBoards.com! Forum Index » FIRE Board
View previous topic :: View next topic  
Author Message
ataloss
**** Heavy Hitter


Joined: 25 Nov 2002
Posts: 559

PostPosted: Mon Jul 21, 2003 5:05 pm    Post subject: Frequent Responses to Hocus- FRH-1- Invalidity Reply with quote

I start off with a simple premise. No matter how smart, Bernstein cannot know what withdrawal rate will be safe in the future. Hocus writes some 1000 word post and I end up responding to various points. Before I know it I am arguing with hocus about how Bernstein got his 2% number (neither of us knows since Bernstein doesn't say in the book.) In point of fact, I don't really care much since my original point was that however smart he is and however sophisticated his approach Bernstein can only make reasoned estimates and projections about the future. Anyway, I am not interested in responding to everything hocus says but I thought I could make an all purpose post to respond to one issue- invalidity of historical safe withdrawal rate studies. Hocus had insisted that we call this the "traditional" approach but I have abandoned his guidance in this matter. This is intended to provide an alternative the hocus position on this matter. Since hocus is accusing me of misrepresenting his I have quoted him and provided dates. I hope gummy doesn't mind being quoted. I am sure that there are many other responses to hocus (but I am too lazy to look for them.)

suggestions, links, corrections (except hocus) are welcome


**************************


Historical swr studies and inflation adjusted safe withdrawal rates

The rate is expressed relative to initial portfolio value. Withdrawals in subsequent years are adjusted upward for inflation.

There are two general approaches to estimating "safe" withdrawal rates from retirement portfolios.

Historically based studies have looked at various portfolio compositions and based on the amount initially withdrawn have reported the probability of portfolio survival. (Example Trinity) or the maximum "100% Safe" withdrawal rate. (REHP) These studies have limitations. If planned withdrawals start at a period of overvaluation more extreme than occurring during the course of the study questions arise about the applicability of the historical swr. Interesting attempts to test the historical swr have included rearranging the yearly data, increasing the expense ratio to simulate possible lower future investment returns, and simply reducing "swr" in a linear fashion according to degree of overvaluation/ historical extremes.

Monte Carlo analysis allows computer simulation of safe withdrawal rates based on estimates of portfolio return and variability. One limitation of this type of study is that future portfolio returns and volatility are unknown.

Some people (ok, hocus) have said that historically based swr analysis is invalid.


Quote:
Please explain how the Trinity study is invalid.

The Trinity study methodology does not take into account how the SWR changes with changes in valuation levels.
Hocus 6/3/03


Note that this does not depend on the rationale of extreme valuation for invalidity. Not all of us agree:


Quote:
About the Trinity Study:
It shows what would have happened had you invested this way or that, in the past.
It doesn't make any claims about the future.
How can it possibly be "invalid"? (Unless, of course, their data was invalid.)

Are these also "invalid" studies?
http://home.golden.net/~pjponzo/sensible-1.gif
http://home.golden.net/~pjponzo/Losses-SPa.gif
Gummy 6/5/03


On the other hand, you can just define classes of studies as invalid:
Quote:
The reason why I say that the 2.0 percent number is valid and the 2.3 percent number is valid, but the 4.0 percent number is invalid, is that the first two numbers both are the product of methodologies which consider the effect of changes in valuation, while the latter is not.
Hocus 7/19/03


Some of us wasted effort trying to tell hocus that changes in valuation occurred during the historical period and were implicit in the study. A waste of effort since he has defined valid studies as those that make explicitly use of valuation.


I thought this was an interesting scattergram of maximal withdrawal rates vs. P/e ratio:
http://rehphome.tripod.com/pestudy1.html
Apparently even devoted followers of the historical approach are aware that there is an association between valuation, investment results, and safe withdrawal rates.
(Thanks to Intercst and BenSolar)

Alternatives to this type of withdrawal include:
Annual withdrawal of a fixed proportion of the portfolio
Preservation of capital with spending dividends/ income
Multiple sub-portfolios/ dynamic asset allocation



_________________
Have fun.

Ataloss


Last edited by ataloss on Wed Aug 20, 2003 6:56 am; edited 2 times in total
Back to top
View user's profile Send private message
[KenM]
*** Veteran


Joined: 04 Mar 2003
Posts: 133

PostPosted: Mon Jul 21, 2003 5:37 pm    Post subject: Reply with quote

A suggested additional quote to show hocus has a rational concept of SWR analysis ....
Quote:
When it comes to SWRs, I see four big factors that are at play: (1) nominal returns; (2) inflation; (3) volatility; and (4) changes in valuation. My view is that the fourth is roughly as important a factor as the other three, and that it would be invalid to do a SWR study ignoring any of the four
Hocus 20 July 2003.
Problem is that he is irrational when he states that historical studies ignore valuation data and therefore are invalid.



_________________
KenM
Never try to teach a pig to sing. It wastes your time and annoys the pig.
Back to top
View user's profile Send private message
JWR1945
***** Legend


Joined: 26 Nov 2002
Posts: 1697
Location: Crestview, Florida

PostPosted: Mon Jul 21, 2003 11:31 pm    Post subject: Reply with quote

ataloss is wrong on a variety of issues! By Definition! Yes, By Definition!

This is a time of intense passions, raw emotions, rising temperatures and sky high blood pressures! We are talking about...dare I say it?...Yes, I will!...semantics! Yes, we can reopen the SWR Definition Summary thread. We can engage in extensive, drawn out discussions about the meaning of words and their uses. There is nothing like semantics to get your juices flowing and to put a spring in your step. It is even better than listening to a software briefing, presented in monotone, in a warm office right after lunch.

[Lord, help us if somebody fails to see the humor in these statements.]

We really can open up the semantics thread at any time. Read the Summary Thread: SWR Definition from Sunday, July 13, 2003 at 4:23 pm CDT.
http://nofeeboards.com/boards/viewtopic.php?t=1107

ataloss has confused the definition of Safe Withdrawal Rates with their context and applications. It is quite difficult to come up with a useful and precise definition. It is critically important to avoid a variety of pathological cases. The points of contention have already been addressed. Passionate debate really should be redirected to definitions and away from content.

And, yes, it would be helpful if hocus and others referred to the definition thread frequently. It is necessary to make some very fine grain distinctions to avoid talking past each other.

Have fun.

John R.


Back to top
View user's profile Send private message
ataloss
**** Heavy Hitter


Joined: 25 Nov 2002
Posts: 559

PostPosted: Tue Jul 22, 2003 3:04 am    Post subject: Reply with quote

Hi JWR, I think we largely agree.

Sometimes hocus refers to the rehp study being flawed and giving a falsely high swr for 2000. I agree. I think that this is something that we can discuss and there may more of less be a consensus that this is the case.

When he refers to the rehp study, trinity and others as being invalid this is strictly a definitional issue. Even if we could somehow prove that a 4% withdrawal rate starting in 2000 would not fail, hocus would still maintain that the studies are invalid.

Chips may say that besides his pension he spends only the dividends from his mixed stock/bond fund portfolio amounting to 2% (I forget the details) This may be safe but by definition hocus would say that it is not a swr. (Because it is not explicitly based on valuation.)

I may be highly risk averse and propose a 1% withdrawal rate from my diversified portfolio. In the interest of consistency hocus would have to say that this is not a "safe withdrawal rate" although he may agree that it is safe lacking an explicit consideration of valuation it would not be a swr by his definition.

I have decided to use the conventional definitions of words. By withdrawal rate I will mean inflation adjusted based on the original portfolio value. By safe I will mean deemed to be unlikely to fail.

I find it paradoxical that if consistent in his use of the term hocus will be calling 2.3% (or whatever) a safe withdrawal rate but will maintain that 1% is not a safe withdrawal rate.
Quote:
ataloss has confused the definition of Safe Withdrawal Rates with their context and applications.


Here we differ. I maintain that the trinity study, rehp study and others were valid (that is they contain no errors) but that with changes in expected returns mean that we should not blindly apply these historically safe withdrawal rates to the future.

You and hocus have defined the swr as a prediction about the future. Thus if intercst had wanted to make a distinction between the result of his study and a prediction he would have had to say something like "a withdrawal rate of x% was safe but the safe withdrawal rate is y%." This seems rather convoluted and unnecessary to me.

Further, I am not sure that hocus is accepting the definitions from the thread since on 5/30 he said this in response to a poll I had put up about a definition:


Quote:
I don't believe that agreeing on a definition will help us resolve this matter.

I believe that the key is that we develop a consensus to make the SWR tool as valuable a tool to aspiring early retirees as it can possibly be. If we have that, all the definitional issues will fall into place, in my view.


I hope you understand that since hocus accuses me of distorting his meaning I prefer to quote him directly.

Perhaps he will post a glossary.



_________________
Have fun.

Ataloss
Back to top
View user's profile Send private message
ataloss
**** Heavy Hitter


Joined: 25 Nov 2002
Posts: 559

PostPosted: Tue Jul 22, 2003 3:19 am    Post subject: Reply with quote

Historical swr studies and inflation adjusted safe withdrawal rates

The rate is expressed relative to initial portfolio value. Withdrawals in subsequent years are adjusted upward for inflation.

There are two general approaches to estimating "safe" withdrawal rates from retirement portfolios.

Studies based on historical sequences of returns have looked at various portfolio compositions and based on the amount initially withdrawn have reported the probability of portfolio survival. (Example Trinity) or the maximum "100% Safe" withdrawal rate. (REHP) These studies have limitations. If planned withdrawals start at a period of overvaluation more extreme than occurring during the course of the study questions arise about the applicability of the historical swr. Interesting attempts to test the historical swr have included rearranging the yearly data, increasing the expense ratio to simulate possible lower future investment returns, and simply reducing "swr" in a linear fashion according to degree of overvaluation/ historical extremes.

Monte Carlo analysis allows computer simulation of safe withdrawal rates based on estimates of portfolio return and variability. One limitation of this type of study is that future portfolio returns and volatility are unknown.

Some people (ok, hocus) have said that historically based swr analysis is invalid.


Quote:

Please explain how the Trinity study is invalid.

The Trinity study methodology does not take into account how the SWR changes with changes in valuation levels.
Hocus 6/3/03



Note that this does not depend on the rationale of extreme valuation for invalidity. Not all of us agree:
Quote:
I don't agree with this claim. In science of all sorts, researchers frequently chose to study one part of a problem and basically ignore everything else. I see it all the time in food and health studies. Just because a study doesn't present the whole picture doesn't mean it is invalid. Just that it is limited. Authors usually include some wiggle words to indicate that they haven't presented the entire picture. The REHP study does that. Bensolat 7/20/03




Quote:
:
About the Trinity Study:
It shows what would have happened had you invested this way or that, in the past.
It doesn't make any claims about the future.
How can it possibly be "invalid"? (Unless, of course, their data was invalid.)

Are these also "invalid" studies?
http://home.golden.net/~pjponzo/sensible-1.gif
http://home.golden.net/~pjponzo/Losses-SPa.gif
Gummy 6/5/03



On the other hand, you can just define classes of studies as invalid:
Quote:
:
The reason why I say that the 2.0 percent number is valid and the 2.3 percent number is valid, but the 4.0 percent number is invalid, is that the first two numbers both are the product of methodologies which consider the effect of changes in valuation, while the latter is not.
Hocus 7/19/03



Some of us wasted effort trying to tell hocus that changes in valuation occurred during the historical period and were implicit in the study. A waste of effort since he has defined valid studies as those that make explicitly use of valuation.


I thought this was an interesting scattergram of maximal withdrawal rates vs. P/e ratio:
http://rehphome.tripod.com/pestudy1.html
Apparently even devoted followers of the historical approach are aware that there is an association between valuation, investment results, and safe withdrawal rates.
(Thanks to Intercst and BenSolar)

Alternatives to this type of withdrawal include:
Annual withdrawal of a fixed proportion of the portfolio
Preservation of capital with spending dividends/ income
Multiple sub-portfolios/ dynamic asset allocation
sensible withdrawal rates
http://home.golden.net/~pjponzo/sensible_withdrawals.htm



_________________
Have fun.

Ataloss
Back to top
View user's profile Send private message
JWR1945
***** Legend


Joined: 26 Nov 2002
Posts: 1697
Location: Crestview, Florida

PostPosted: Tue Jul 22, 2003 5:22 am    Post subject: Reply with quote

I strongly disagree with ataloss. By definition!

ataloss clearly rejects our standard definition, which we made to avoid distractions based on semantics. For example:
Quote:
Chips may say that besides his pension he spends only the dividends from his mixed stock/bond fund portfolio amounting to 2% (I forget the details) This may be safe but by definition hocus would say that it is not a swr. (Because it is not explicitly based on valuation.)

It is not what hocus would say. It is what the standard definition says.
Quote:
I have decided to use the conventional definitions of words. By withdrawal rate I will mean inflation adjusted based on the original portfolio value. By safe I will mean deemed to be unlikely to fail.

We cannot avoid endless confusion because we are not talking about the same things.

Here is the link to the Summary Thread: SWR Definition and my first post on that thread:
http://nofeeboards.com/boards/viewtopic.php?t=943
http://nofeeboards.com/boards/viewtopic.php?p=6859#6859
This is our official definition:
Quote:
The Safe Withdrawal Rate is the best estimate of the maximum withdrawal amount for a retirement portfolio based on both a well-defined procedure and a set of well-defined constraints. It is stated as a percentage of the initial portfolio balance. It must be a mathematical calculation that is based primarily on existing, historical information. The sense in which it is the best estimate must be identified. The algorithm or procedure must be stated clearly. The constraints must be stated clearly. There must be an assessment of the degree of safety. This may be stated in terms of probabilities.

The phrase Safe Withdrawal Rate should always be placed in context. It should be defined in terms of those factors that are covered by the calculation, those factors that are not covered and those that are only partially covered. The context of a Safe Withdrawal Rate calculation must always address the reliability of the estimate. This should include known sensitivities to the assumptions that are inherent in making any projection. It is desirable to include general comments about the applicability of the estimate.

The Safe Withdrawal Rate must be described in terms independent of any methodology. The Safe Withdrawal Rate must be based on calculations that are independent of the actual occurrence of future events. The Safe Withdrawal Rate must be described in terms that acknowledge our inability to conduct carefully controlled experiments.

I amplified it with this, taken from:
http://nofeeboards.com/boards/viewtopic.php?p=8382#8382
Quote:
My approach is different. My approach is to restrict the term Safe Withdrawal Rate to being as objective as possible. My approach is to define it as a mathematical calculation which, at least in theory, has a right answer. To do so required me to allow for the single phrase Safe Withdrawal Rate to apply to a multitude of different questions. As such, the term must always be placed in its context: What question does the mathematical calculation attempt to answer? How well does it do that? To what extent can the calculation be applied to similar, but different, questions?

Each individual then constructs his own rules of thumb, based upon a variety of such calculations and according to his own personal applications.

I have defined Safe Withdrawal Rate in such a manner that all valid mathematical computations are retained even when there are known, significant shortcomings. It is in this context that I have retained all of the studies that fail to include valuations as a key factor. Through extensive research, we have firmly established that valuations always matter and that valuations have been highly significant in recent years. I have approached this issue in terms of accepting the older results as is and then making adjustments for today. It must be understood that recognizing the true importance of valuations is a new finding. Earlier studies are not to be faulted for excluding valuations. They were assumed to be of secondary importance and that assumption proved false. The correct action at this point is to build upon and correct the previous studies, not to toss them aside.

Here is where ataloss first rejected the official definition:
http://nofeeboards.com/boards/viewtopic.php?p=8398#8398
Quote:
I guess I am with raddr (per the quote) until I see this objective swr I am withholding judgment. "As objective as possible" sounds subjective to me.

I haven't read the Ben Stein book but I think that it is common knowledge that one could do better (and by better I mean having higher returns)
buying stocks when they are cheap. Market timing (like other investment approaches) looks easy in retrospect but may be difficult to implement prospectively.

Here is hocus's overall position, stated well.
http://nofeeboards.com/boards/viewtopic.php?p=8389#8389

If we need to open up the SWR Definition thread again, that's OK. But let us stop going around and around in circles simply because of definitions.

Have fun.

John R.


Back to top
View user's profile Send private message
hocus
Moderator


Joined: 02 Dec 2002
Posts: 435

PostPosted: Tue Jul 22, 2003 5:55 am    Post subject: Reply with quote

The correct action at this point is to build upon and correct the previous studies, not to toss them aside.

I very strongly agree with this statement.

Through extensive research, we have firmly established that valuations always matter and that valuations have been highly significant in recent years

I very strongly agree with this statement.

I have defined Safe Withdrawal Rate in such a manner that all valid mathematical computations are retained even when there are known, significant shortcomings.

I very strongly disagree with this statement.

It might be helpful if someone could track down a statistics textbook or a text on how to do research and see if we can locate a well-accepted description of what is required of a methodology for it to be considered "valid" in the scientific community.


Back to top
View user's profile Send private message
BenSolar
*** Veteran


Joined: 25 Nov 2002
Posts: 242
Location: Western NC

PostPosted: Tue Jul 22, 2003 6:44 am    Post subject: Reply with quote

I agree with JWR that the solution to the recent quagmire is in argeeing to and relying upon a definition. Otherwise we talk past each other using different definitions. That is no fun.
hocus wrote:
[JWR1945 wrote: ]I have defined Safe Withdrawal Rate in such a manner that all valid mathematical computations are retained even when there are known, significant shortcomings.

I very strongly disagree with this statement.


From our definition:
Quote:
The constraints must be stated clearly. There must be an assessment of the degree of safety. This may be stated in terms of probabilities.

The phrase Safe Withdrawal Rate should always be placed in context. It should be defined in terms of those factors that are covered by the calculation, those factors that are not covered and those that are only partially covered. The context of a Safe Withdrawal Rate calculation must always address the reliability of the estimate. This should include known sensitivities to the assumptions that are inherent in making any projection. It is desirable to include general comments about the applicability of the estimate.

Thus the REHP style SWR would fall under the definition only when accompanied by notice that valuation isn't explicitly considered, and discussion on the possible effects of that exclusion.

I think this is entirely reasonable. I think, given the state of knowledge about the effect of valuation, that the currently used wiggle words of 'if the future (worst case stock market return) is no worse than the past' and ' '100% safe' ' (in quotations to mean that it passed 100% of the historical periods) are not really sufficient notice of the limitations. Though they do provide enough wiggle room that defenders can typically hide behind them.

They can be made to look a bit silly though, as they end up claiming that the study has no utility for predicting the future. If that's the case, why bother looking at it at all? :DWe look at the past to determine our best course for the future, end of story. We have no other guide except logic based on unproved assumptions. To throw up one's hands and say 'no one can predict the future' is to throw away almost every tool we have for financial planning.

Finally, ataloss, in his FRH, transforms the REHP style 'SWR' from pretty much failing to meet our definition, to meeting it with these phrases:
Quote:
These studies have limitations. If planned withdrawals start at a period of overvaluation more extreme than occurring during the course of the study questions arise about the applicability of the historical swr.

That's pretty much all it takes: voila, the REHP style studies are transformed into valid SWR studies according to our definition. Or at least well on their way. They may still be missing nicities, but the essentials are there. They may not account for undervaluation, but the number is certainly still safe as valuation drops. I would add a caveat that at the higher levels of valuation seen in the past, there may still be considerable danger from the possibility of a worse return sequence than that seen from previous valuation peaks.

Regards,



_________________
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
Back to top
View user's profile Send private message
hocus
Moderator


Joined: 02 Dec 2002
Posts: 435

PostPosted: Tue Jul 22, 2003 7:41 am    Post subject: Reply with quote

Thus the REHP style SWR would fall under the definition only when accompanied by notice that valuation isn't explicitly considered, and discussion on the possible effects of that exclusion. I think this is entirely reasonable.

I understand that this is where you and JWR1945 are coming from, BenSolar. The two of you (and probably raddr too and perhaps a few others) are saying "yes, changes in valuation levels always affects the SWR, but if you note in the text of the study the failure to consider this effect, you have put readers on notice that you are not calculating the SWR as it is defined for purposes of SWR analysis." It's a big step forward to ask of researchers that they do this. I don't find fault with the idea. I don't think it goes far enough, however.

What JWR1945 says (if I read him right) in his statement above is that any SWR study containing a correct mathematical calculation somewhere within it is for that reason declared "valid." I have the greatest respect possible for JWR1945, but I think that's a silly standard for us to apply.

A "researcher" cuts up 99 slips of paper and writes the nunmbers1 through 99 on them, puts them in a hat, and then pulls three of them out. He adds up the three numbers--2, 7, and 14. He issues a study stating that "The historical data reveals that you can take a withdrawal of 23 percent of your portfolio with 100 percent certainty that the portfolio will survive 30 yeasrs, presuming that the future is like the past."

Valid study? I say no. <>b>JWR1945 says yes because there is indeed a correct mathematical calculation performed under this methodology. The numbers 2, 7, and 14 were added together properly to get the SWR of 23. It's all complete and utter nonsense. But it's valid, by our proposed definition.

I am saying that we need to set the bar a little higher than that. We know that changes in valuation levels always affect SWRs, and we know that the effect is in some circumstances an effect that causes the studies to be $1 million or more off in their results. Not because the researchers did anything wrong re the math. The math is perfect.. It's the methodology that is the problem. You can't get there from here. You cannot determine the SWR without taking into account the critical factor of changes in valuation levels.

Let's consider a practical question. Does it matter? Yes! It matters a bunch.

There are three SWR threads going on over at the TMF board today. Take a look at them and tell me what you see. Do you see evil people trying to mislead others and cause busted retirements? Do you see stupid people making a hash of things? That's not what I see. I see highly confused people, trying to make sense of something and failing utterly.

These people want to know what is safe, for obvious reasons. But they have been misled by analysts who claim to know what they are talking about but obviously do not. I don't mean intercst. There is a sense in which intercst is a victim here. He believed in this conventional methodology. One of the guys who came up with it graduated from Harvard University, didn't he? That's cause to trust that the methodology is aimed at providing a reasonable answer to the question posed, isn't it?

This methodology does not provide reasonable answers to the question posed. They are always wrong, and sometimes they are way, way off the mark. I don't care whether the guy who thought this stuff up graduated from Harvard University or from Joe's Barber School. He's got to get about the business of fixing what he did, or some others need to step in and fix it for him.

People are seeing their retirements go down in flames because they have put their trust in this stuff and it is all just completely wrongheaded. These studies don't tell you what the historical data says. They tell you what this crazy methodology says. That's not at all the same thing.

I have no complaint with the Harvard guy if, after reading the report I produce, he comes here and tells us, "Oh, that's so interesting, I never realized that before." That's just fine. But if this Harvard individual comes here and says "Yes, I see that the methodology always produces the wrong answer, but I am just going to continue producing studies that report wrong answers because it is too much trouble for me to go about making changes," he's going to hear the old what-for from me. I do not say the sort of things I am saying here behind people's backs and then get all shy and quiet when they are present in the room. I will tell him to his face what I think about what he has done to the Retire Early dreams of good people who counted on him to provide reasonable advice on an important money question.

My presumption is that the guy from Harvard is a man of good will. That is always my presumption. It may be that I will become fast friends with him because he will be impressed that I discovered the flaw in the conventional SWR methodology and led a campaign to have it declared invalid. There's no reason to believe that it will not turn out that way, is there? That is certainly how I hope it turns out.

But I do not get this thing where we conclude that the results of the methodology are always wrong, but people may continue to issue studies based on it knowing darn well that it causes busted retirements. The people who use these studies cannot be expected to examine every assumption in them to the extent we have and figure out whether they are valid or not. There have to be standards applied to people putting forward their research for use by the public. Researchers have responsibilities like everyone else and one of them is to consider whether the methods they are using to answer questions are capable of providing accurate answers to those questions.

That's where I am coming down on this, anyway. I am less worried about the researchers and more concerned about the people who make use of the research. I think the people making use of it have a right to expect that reasonable methodologies be employed. This one is not reasonable. It was reasonable at one time. Not anymore. Not after what we have discovered about it from the research that has been developed by this board community.


Back to top
View user's profile Send private message
ataloss
**** Heavy Hitter


Joined: 25 Nov 2002
Posts: 559

PostPosted: Tue Jul 22, 2003 7:42 am    Post subject: Reply with quote

Quote:
It might be helpful if someone could track down a statistics textbook or a text on how to do research and see if we can locate a well-accepted description of what is required of a methodology for it to be considered "valid" in the scientific community.


I would be interested in any statements from a statistical book that indicate that you can automatically declare invalid a type of study. That is, can you really say "I don't care if the REHP study results are correct, the study is invalid." I admit it has been a while since I took any statistics classes.


ataloss
Quote:
Chips may say that besides his pension he spends only the dividends from his mixed stock/bond fund portfolio amounting to 2% (I forget the details) This may be safe but by definition hocus would say that it is not a swr. (Because it is not explicitly based on valuation.)


jwr
Quote:
It is not what hocus would say. It is what the standard definition says.


I am not supporting any definition that suggests that raddr's 2.3% number is a valid swr if it requires that chips 2% rate is not a safe withdrawal rate. It is safe(er) and it is a withdrawal rate but it is not a safe withdrawal rate? Until the trademark takes effect I will continue to use the conventional meanings of words. Wink
Quote:

I strongly disagree with ataloss. By definition!

LOL :)Join the club. ES has disagreed with me about international markets for years.

JWR, My object isn't to have hocus change his mind. I just want to suggest an alternative viewpoint. In the past I kept trying to get hocus to "see the light" without success. Now I am perfectly happy presenting my version of the facts. If you hocus and others want to define terminology I have bo objection at all.



_________________
Have fun.

Ataloss
Back to top
View user's profile Send private message
BenSolar
*** Veteran


Joined: 25 Nov 2002
Posts: 242
Location: Western NC

PostPosted: Tue Jul 22, 2003 8:26 am    Post subject: Reply with quote

hocus wrote:
I understand that this is where you and JWR1945 are coming from, BenSolar. The two of you (and probably raddr too and perhaps a few others) are saying "yes, changes in valuation levels always affects the SWR, but if you note in the text of the study the failure to consider this effect, you have put readers on notice that you are not calculating the SWR as it is defined for purposes of SWR analysis."


No, I'm saying that if you note the limitation, then "you are calculating the SWR as it is defined for purposes of SWR analysis." At least using the definition we've come up with so far.

You are operating on a different definition than everyone else, and it makes constructive dialog next to impossible.



_________________
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
Back to top
View user's profile Send private message
BenSolar
*** Veteran


Joined: 25 Nov 2002
Posts: 242
Location: Western NC

PostPosted: Tue Jul 22, 2003 8:29 am    Post subject: Reply with quote

hocus wrote:
A "researcher" cuts up 99 slips of paper and writes the nunmbers1 through 99 on them, puts them in a hat, and then pulls three of them out. He adds up the three numbers--2, 7, and 14. He issues a study stating that "The historical data reveals that you can take a withdrawal of 23 percent of your portfolio with 100 percent certainty that the portfolio will survive 30 yeasrs, presuming that the future is like the past."

Valid study? I say no. <>b>JWR1945 says yes because there is indeed a correct mathematical calculation performed under this methodology. The numbers 2, 7, and 14 were added together properly to get the SWR of 23. It's all complete and utter nonsense. But it's valid, by our proposed definition.


This is absurd, and wouldn't remotely fall under the definition. The several paragraphs of language JWR put together encompass quite a bit more than 'a mathematical calculation'.



_________________
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
Back to top
View user's profile Send private message
BenSolar
*** Veteran


Joined: 25 Nov 2002
Posts: 242
Location: Western NC

PostPosted: Tue Jul 22, 2003 8:33 am    Post subject: Reply with quote

BenSolar wrote:
This is absurd, and wouldn't remotely fall under the definition. The several paragraphs of language JWR put together encompass quite a bit more than 'a mathematical calculation'.


An important part of JWR's definition left unfullfilled in your absurd example:
Quote:
It must be a mathematical calculation that is based primarily on existing, historical information.


Drawing random numbers is not based on historical information.



_________________
"Do not spoil what you have by desiring what you have not; remember that what you now have was once among the things only hoped for." - Epicurus
Back to top
View user's profile Send private message
JWR1945
***** Legend


Joined: 26 Nov 2002
Posts: 1697
Location: Crestview, Florida

PostPosted: Tue Jul 22, 2003 10:25 am    Post subject: Reply with quote

To ataloss:

I must insist that you restrict yourself to the official definition if you are going to use the term Safe Withdrawal Rate. It is OK for you to talk about what you are talking about. Just use different words. You may work on changing the official definition if you so desire. That is OK. But until you do, I must insist that you restrict your choice of words. Otherwise, all communications cease. My own words become meaningless. Progress becomes impossible.

hocus's reservations about the official definition of the Safe Withdrawal Rate is not new. His most recent explanation is excellent. Up until now, he has been willing to make a concession to us in order to advance our knowledge. After all, if everybody plays by the same rules, it would be easy enough for hocus to write an explanation of his reservations and to offer guidance to readers.

The substantive issue behind hocus's concern has to do with how you handle previous reports that have deficiencies that we know about now, but that we did not know about when they were written. We decided to retain those reports, provided that their deficiencies and the implications thereof were spelled out clearly as part of their context. hocus, himself, provided an example previously that involved maps generated by people who thought that the earth was flat. Under some circumstances, primarily for local maps, they might still be useful even though they are known to be wrong. For example, they might be the best maps for locating caves. Or old cemeteries.

It might be best to describe such maps as invalid but usable. But when it comes to Safe Withdrawal Rate investigations, it is more difficult because different studies have different strengths and weaknesses. Using the map analogy, a local map might still be the best map in existence for a particular purpose. The issue becomes more difficult if you state strongly that the map is invalid.

It is critically important that we mean the same thing when we use the same words.

Have fun.

John R.


Back to top
View user's profile Send private message
[KenM]
*** Veteran


Joined: 04 Mar 2003
Posts: 133

PostPosted: Tue Jul 22, 2003 6:46 pm    Post subject: Reply with quote

I have no problems understanding the way ataloss defines SWRs in the context of this Frequent Response to Hocus - to me, the language of an Official Definition is equivalent to a legal ordinance - fine for use in court but not always used in everyday conversation.
Quote:
The issue becomes more difficult if you state strongly that the map is invalid.
......and this is one of the major problems - hocus has chosen the "invalid" word very carefully - it's not a question of semantics.
The substance of his conclusion is fine - it's all the incorrect stuff he appears to insist upon packaging with that conclusion in his crusade against REHP that, regrettably, draws attention away from the conclusion and consequently undermines the substance of the conclusion.



_________________
KenM
Never try to teach a pig to sing. It wastes your time and annoys the pig.
Back to top
View user's profile Send private message
[KenM]
*** Veteran


Joined: 04 Mar 2003
Posts: 133

PostPosted: Tue Jul 22, 2003 6:49 pm    Post subject: Reply with quote

Ataloss
Perhaps another quote.........
Quote:
hocus, have you actually read any of the original studies?

I read the intercst study when I first came to the TMF board in 1999. I haven't read any others since.
Hocus 19 July 2003



_________________
KenM
Never try to teach a pig to sing. It wastes your time and annoys the pig.
Back to top
View user's profile Send private message
hocus
Moderator


Joined: 02 Dec 2002
Posts: 435

PostPosted: Wed Jul 23, 2003 12:55 am    Post subject: Reply with quote

hocus has chosen the "invalid" word very carefully - it's not a question of semantics.
The substance of his conclusion is fine

I think it is helpful to the entire board community for you to acknowledge that, KenM..

it's all the incorrect stuff he appears to insist upon packaging with that conclusion in his crusade against REHP that, regrettably, draws attention away from the conclusion

If you check the posts, you will find that it is not me who brought up the REHP matter. It is my preference that it not be brought up. I think that the better course of action is to deal first with the substance question. Any process issues are better put off for some other time, in my view.


Back to top
View user's profile Send private message
ataloss
**** Heavy Hitter


Joined: 25 Nov 2002
Posts: 559

PostPosted: Wed Jul 23, 2003 1:43 am    Post subject: Reply with quote

JWR, I hope you like this version better. Reviewing the old threads trying to figure out what hocus was talking about I enjoyed looking at your posts. Since we can't make people coming to this board sign a pledge to accept a particular definition of swr I thought I should start by using words according to conventional meanings, show what hocus has said, then refer to the official definition. Maybe there should be a link. I have not incorporated anything about the hocus post yesterday which seems suggest a complete misunderstanding of the nature of historical swr studies. My goal is that unless they wish to waste their time people should stop making points to hocus that have been made many times on this board and elsewhere. If there are no further comments or corrections I plan to drop the matter of invalidity and FRH for the time being. If in 6 months hocus mentions invalidity and I see people getting confused I will repost. I am not seeking approval from hocus for this post and I am not trying to change his mind. He has his viewpoint and I have mine. If anyone else wants to copy part of all of this first FRH to respond to hocus they are welcome to do so.

*****************

Historical swr studies and inflation adjusted safe withdrawal rates

The rate is expressed relative to initial portfolio value. Withdrawals in subsequent years are adjusted upward for inflation.

There are two general approaches to estimating "safe" withdrawal rates from retirement portfolios.

Studies based on historical sequences of returns have looked at various portfolio compositions and based on the amount initially withdrawn have reported the probability of portfolio survival. (Example Trinity) or the maximum "100% Safe" withdrawal rate. (REHP) These studies have limitations. If planned withdrawals start at a period of overvaluation more extreme than occurring during the course of the study questions arise about the applicability of the historical swr. Interesting attempts to test the historical swr have included rearranging the yearly data, increasing the expense ratio to simulate possible lower future investment returns, and simply reducing "swr" in a linear fashion according to degree of overvaluation/ historical extremes.

Monte Carlo analysis allows computer simulation of safe withdrawal rates based on estimates of portfolio return and variability. One limitation of this type of study is that future portfolio returns and volatility are unknown.

Using a particular definition of "safe withdrawal rates" (swr) some people (mainly hocus) have argued that historically based swr analysis is invalid. Most people seem to use the term "safe withdrawal rates" as if it meant withdrawal rates which are safe. Hocus has a specialized meaning in mind:


Quote:
Quote:
Please explain how the Trinity study is invalid.

The Trinity study methodology does not take into account how the SWR changes with changes in valuation levels.
Hocus 6/3/03




And here by failure to account for changes in valuation, he means failure to explicitly include valuation in some sort of calculation.. Many of us, using the conventional meanings of words have disagreed.
Quote:
Quote:
I don't agree with this claim. In science of all sorts, researchers frequently chose to study one part of a problem and basically ignore everything else. I see it all the time in food and health studies. Just because a study doesn't present the whole picture doesn't mean it is invalid. Just that it is limited. Authors usually include some wiggle words to indicate that they haven't presented the entire picture. The REHP study does that. Bensolar 7/20/03





Quote:
Quote:
:
About the Trinity Study:
It shows what would have happened had you invested this way or that, in the past.
It doesn't make any claims about the future.
How can it possibly be "invalid"? (Unless, of course, their data was invalid.)

Are these also "invalid" studies?
http://home.golden.net/~pjponzo/sensible-1.gif
http://home.golden.net/~pjponzo/Losses-SPa.gif
Gummy 6/5/03



Further hocus on this matter:
Quote:
Quote:
:
The reason why I say that the 2.0 percent number is valid and the 2.3 percent number is valid, but the 4.0 percent number is invalid, is that the first two numbers both are the product of methodologies which consider the effect of changes in valuation, while the latter is not.
Hocus 7/19/03




Some of us wasted effort trying to tell hocus that changes in valuation occurred during the historical period and were implicit in the study. A waste of effort since he has defined valid studies as those that make explicitly use of valuation in some sort of calculation.

Some of us wanted to say that perhaps a study could be valid but less applicable to current circumstances. For example at the peak of the market bubble the market's p/e ratio was above the range that existed in the 1870-1970 period studied. Monte Carlo analysis using expected future returns suggested a lower swr. Wasted words since for hocus the "invalidity" of the study is definitional and does not depend in any way on the study actually giving a wrong or unreliable swr estimate. The study is invalid even for the 1870-1970 period where the withdrawal rate is know to have been safe. Hocus has also criticized the study for giving too high a withdrawal estimate and for causing "busted retirements" but I think it is important to understand that the invalidity claim if definitional and is not subject to disproof.



I thought this was an interesting scattergram of maximal withdrawal rates vs. P/e ratio:
http://rehphome.tripod.com/pestudy1.html
Apparently even devoted followers of the historical approach are aware that there is an association between valuation, investment results, and safe withdrawal rates.
(Thanks to Intercst and BenSolar)

Alternatives to this type of withdrawal include:
Annual withdrawal of a fixed proportion of the portfolio
Preservation of capital with spending dividends/ income
Multiple sub-portfolios/ dynamic asset allocation
sensible withdrawal rates
http://home.golden.net/~pjponzo/sensible_withdrawals.ht

There is an "official" swr definition on the NFB FIRE board but hocus defines swr in a somewhat different manner.



_________________
Have fun.

Ataloss
Back to top
View user's profile Send private message
hocus
Moderator


Joined: 02 Dec 2002
Posts: 435

PostPosted: Wed Jul 23, 2003 2:58 am    Post subject: Reply with quote

for hocus the "invalidity" of the study is definitional and does not depend in any way on the study actually giving a wrong or unreliable swr estimate. The study is invalid even for the 1870-1970 period where the withdrawal rate is know to have been safe.

I take a practical perspective on this thing. I am not hung up on definitions. If I were to be persuaded that the conventional methodology is invalid from a theoretical perspective, but that the future practical consequences of the invalidity were not likely to be great, I would be inclined to put my energies in some other direction.

I am 99 percent sure that the conventional methodology provided bad numbers during the bubble period, and that the consequences of that were very bad. People thought they were embarking on safe retirements and in fact they were embarking on high-risk retirements. On that issue, I think the book is closed.

If that were the only issue, I am not sure whether I would pursue this or not. We can't go back and change what happened in the year 2000. It's over. If I thought that there were no reason to expect this invalidity problem to cause significant harm to aspiring early retirees in the future, then I might well drop it. But I am not at all convinced that that is the case.

For one thing, it is possible that we could return to bubble valuation levels at some point in the future. My understanding is that we may even be over the line today. If we are not, we are pressing right up against it. So the concern over invalidity at bubble-level valuations is not a dead one.

Moreover, I am not at all sure that the conventional methodology is valid at pre-bubble valuation levels. I believe that it is invalid at all valuation levels. I don't possess the same confidence on this issue as I do on the bubble years issue. But the data I have seen leads me to believe that the invalidity problem is an ever-present one.

It is on the question of pre-bubble year validity that the board community could give me the most help. I don't want to write an article saying "I think the conventional methodology is invalid even in the pre-bubble years." I want to be able to say either: (1) "the conventional methodology is invalid even in the pre-bubble years"; or (2) the problem is confined to the bubble years and does not extend to time-periods in which valuation levels are within the range examined in the studies.

I am not prepared at this time to write either sentence in an article aimed at a general audience. I lean strongly toward the first sentence. But I would like to be absolutely sure before venturing too much farther out on the limb that I am on.


Back to top
View user's profile Send private message
[KenM]
*** Veteran


Joined: 04 Mar 2003
Posts: 133

PostPosted: Wed Jul 23, 2003 3:14 am    Post subject: Reply with quote

Quote:
I think it is helpful to the entire board community for you to acknowledge that, KenM

Playing the mis-representation game again , hocus Question
Now I know why few people are sympathetic with your views Mad



_________________
KenM
Never try to teach a pig to sing. It wastes your time and annoys the pig.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    NoFeeBoards.com! Forum Index » FIRE Board All times are GMT - 9 Hours
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001- 2004 phpBB Group
Designed for Trushkin.net | Styles Database