Any Studies on Long-term Market Timing?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy » Fri Jan 28, 2011 10:52 am

cjking wrote:
DRiP Guy wrote:Here is the most recent post of a proponent of using analysis of valuation and other measures to time the market, who put his money where his mouth is. (Well, actually other people's money, but that's a minor nit.)

I highly recommend following his entire history of posts as a terrific exemplar of behavioral difficulties intrinsic to any market timing attempts, even those rooted in the most carefully developed and strenuously tested approaches.


I see no connection between any strategy MarketTimer has ever documented on this site and what is being debated in this thread. This thread is about timing, but in relation to (a) one particular asset class (b) one particular valuation mechansim and (c) one particular strategy for altering equity exposure on the basis of the valuation mechanism. Change any of those elements and there's a very high probability that those on the opposite side if the argument to you in this thread will then be on the same side of the argument as you.


At the risk of seeming flip: Potato, pa-tah-toe.

At the end of the day, behaviorist tendencies tend to undermine even the most elaborately detailed of these timing schemes. I think Mike Tyson was already quoted earlier, but it bears a reprise: "Everyone has a plan until they get hit." If we found that even the most stalwart of Bogleheads, supposedly committed to the most simple of strategies ("do nothing!)", were found wavering when the tides of the market changed a bit, then how much more so for the active trading that timing strategies mandate? The opportunity for rationalization is ever present, due to "new information."

I guess I merely wish to be understood that my skepticism is not so much for PE10 as a metric, it is for the espousing of *any* scheme of trying to apply an algorithmic trading strategy to boost returns above market levels. As I believe Bernstein points out, the relentless rules of arithmetic mandate that such schemes on the whole can be successful only to a degree offset by losers biting the pavement on the other end of such strategies... less the friction of trading costs for all concerned (gotta pay the croupier).

Please don't misunderstand me -- I love to read these ideas, and hope you and others continue to present, debate, and analyze them. However, I find myself compelled to pop in from time to time to recite what I *think* is both personal objective learned truth, as well as board ethos. If I am wrong, I both apologize and stand ready to be corrected.

And if the mere frequency of me doing so is boorish, in and of itself, then I also apologize, but only to the extent I feel that NOT reiterating a difference in opinion tends to leave observers with an impression of agreement having been reached.

crl848
Posts: 6
Joined: Fri Jan 28, 2011 6:23 am

Post by crl848 » Fri Jan 28, 2011 11:50 am

bob90245 wrote:The TAA model under discussion doesn't direct the investor "to get out of the market" when triggered. You merely lower your stock allocation from neutral, like 60%, to conservative, like 30%. So you are always in the market. Only less at high valuations and more at low valuations.
.


Absolutely. Take what I said as shorthand for reducing your equity allocation to whatever your low point is.

As Cliff Asness once said, market timing is hard!

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Jan 31, 2011 9:13 pm

crl848, thanks for your very well thought out comments.

I'm still reading and thinking about the issues of this thread. I'm just providing some partial comments for now.

My copy of Graham and Dodd's 1940 edition of Security Analysis arrived. Thanks DRiP Guy! Here is what they say on page 685 and 686:

Graham and Dodd wrote:Can the analyst exploit successfully the repeated exaggerations of the general market? Experience suggests that a procedure somewhat like the following should turn out to be reasonably satisfactory:
1. Select a diversified list of leading common stocks, e.g., those in the 'Dow-Jones Industrial Average.'
2. Determine an indicated 'normal' value for this group by applying a suitable multiplier to average earnings. The multiplier might be equivalent to capitalizing the earnings at, say, twice the current interest rate on highest grade industrial bonds. The period for averaging earnings would ordinarily be seven to ten years, but exceptional conditions such as occurred in 1931-1933 might suggest a different method ...
3. Make composite purchases of the list when the shares can be bought at a substantial discount from normal value, say, at 2/3 such value. Or purchases may be made on a scale downwards, beginning say, at 80% of normal value.
4. Sell out such purchases when a price is reached substantially above normal value, say, 1/3 higher, or from 20% to 50% higher on a scale basis.


The authors then list various limitations and things, but conclude:

Graham and Dodd wrote:But for those who realize its inherent limitations it may have considerable utility, for at least it is likely on the average to result in purchases at intrinsically attractive levels - which is more than half the battle in common-stock investment.


My comments about this:

1. A definition of PE10 is buried in there, but Graham and Dodd are also suggesting that looking at only PE10 is not enough, as interest rates should be in there somehow too. It's a good point that has been brought up earlier by peter71, and eventually I should update the regression model I used in a previous paper with (100/PE10), dividend yield, and interest rates in order to do this properly.

2. This is a very clear definition of valuation-based tactical asset allocation.

3. The purpose of this is to help investors battle their natural tendencies of buying stocks high and selling them low. In that regard, valuation-based indexing has the same purpose as either buy-and-hold, or periodic rebalancing to an underlying asset allocation. All of these strategies are subject to the same behavioral mistakes: regret bias could cause you to increase your stock allocation just before a peak (such as 1999) and a short-term memory could cause you to abandon your stocks entirely at precisely the wrong time (such as 1982). With a valuation-based approach, the errors would be amplified, as you would be making the error from a more unfortunate starting point, but it is all the same errors. But at the same time, wearethefall made the comment a while back that considering valuations may also allow you to stay the course more firmly with your 'buy and hold' strategy and not make behavioral mistakes.

A few other comments:

1. cjking made a contribution a while back about using a decision rule that doesn't let you change your allocation back to normal levels until the median is crossed again. Actually, in that book by David Jenkins linked to before, this is called the 'halfway rule' and was quite popular with many of the formula plans used in the 1940s and 1950s. It provides an even stronger commitment to the idea of regression to the mean (median) than did my original example.

2. I think I will be able to get P/E data for Japan back to 1956, which will let me construct PE10 since 1966. Japan provides a natural laboratory (which I think qualifies as out-of-sample) to test what happens with these strategies under extreme circumstances.

3. I don't mean sound controversial, and I'm only saying this here where it can be discussed in a rational and pleasant manner by informed investors. I'm not going to go out and shout this on the streets. I understand that it could be confusing and counterproductive for unsophisticated investors to hear if it leads them to market timing. But, I can't help but to think that 'buy and hold' is also a form of long-term market timing, at least to the extent that these valuation-based strategies must share that title. Why? Stocks are risky, but investors who buy and hold stocks are doing so under the belief that over a sufficiently long period, they will be compensated for their risks by receiving higher returns. Otherwise, why hold stocks? All of this is just a matter of degrees: valuations do require further timing decisions than 'buy and hold', but 'buy and hold' does have the timing implications I suggest, which may have been forgotten about in the intervening years since the concept was developed.

4. I'm still waiting to receive Lucile Tomlinson's books so that I can figure out whether she added something beyond Graham and Dodd with her 'compromise plan'.

5. A side discussion came about regarding whether following this strategy can make you become Warren Buffett (I'm overly simplifying that discussion). Of course not. It is more about finding a behavioral rule to help us from making mistakes, precisely because we are not like Warren Buffett.

6. I think I may have sold this strategy the wrong way in my example. I was excited to see so much outperformance, but really perhaps it is the reduced risk side of the story that should be emphasized. cjking made a comment in this regard.

7. I can't completely believe that valuations-based tactical asset allocation violates the tenets of Bogleheads. I understand Alex's point that discussing it can provide a slippery slope that leads one into the clutches of market timing. For that reason, valuations-based investing should not be part of the curriculum of Bogleheads 101. But once you master the basics of Bogleheads 101, can't this topic fit onto the syllabus for Bogleheads 102?

Best wishes, Wade

User avatar
jidina80
Posts: 729
Joined: Wed Feb 24, 2010 5:05 pm
Location: Fiji

Post by jidina80 » Mon Jan 31, 2011 10:11 pm

Doesn't the flow of money into and out of mutual funds provide data for a good long-term timing model? Since investor returns lag behind fund returns, it seems that buying when most investors are selling, and selling when most investors are buying, would allow one to out-perform a fund by approximately the same percentage that the average investor is lagging. Has anyone tried this kind of model?

Just.

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy » Mon Jan 31, 2011 11:21 pm

Congrats Wade! I'd say the prof has been doing his homework! I'm encouraged by your thoughtful approach to carefully analyzing and potentially melding a variety methods from several approaches to come up with what one would hope to be a robust and outperforming process.

So that the absolute sincerity of those remarks will not be tainted, I am going to forgo what has become for me an almost a de rigueur series of comments and cautions about better mousetraps, the EMH, and such.

I honestly applaud your effort and your work, and hope you can uncover insights to help people plan and fund their retirements, with a clearer understanding of risk than is often held today.

As to your plea: "I can't completely believe that valuations-based tactical asset allocation violates the tenets of Bogleheads," I think it is well and often established that Bogle himself did not object to putting your finger on the scale of your allocations,within a framework, based on a holistic (and one hopes accurate) survey of market conditions. I don't begrudge him that, or anyone who espouses it. I do say, however, that I personally have found it easier to understand my own risks and objectives, and to deny the opportunity for myself to trudge down the all-too-often seen behaviorist tendencies to do precisely the wrong thing at precisely the wrong time. I think of it as a measure of candor, open-mindedness and genuine flexibility that he entertained such notions.

Personally, I weight those one or two occasional nods to market valuations, against the consistent, broader, deeper context of within the entire body of work he developed, and what *I* think buy and hold really means, and investing for the long run actually denotes, and I find that such excursions are simply not consistent with my own needs, risk tolerances, or capabilities, nor are they required. Mine is *not* a 'returns optimizing strategy.' It is a probability exercise in being able to fund a modest retirement in a reasonably low risk manner.
Last edited by DRiP Guy on Mon Jan 31, 2011 11:29 pm, edited 1 time in total.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Post by bob90245 » Mon Jan 31, 2011 11:23 pm

wade wrote:7. I can't completely believe that valuations-based tactical asset allocation violates the tenets of Bogleheads. I understand Alex's point that discussing it can provide a slippery slope that leads one into the clutches of market timing. For that reason, valuations-based investing should not be part of the curriculum of Bogleheads 101. But once you master the basics of Bogleheads 101, can't this topic fit onto the syllabus for Bogleheads 102?

Not sure if you have Jack Bogle's view on this from his book Common Sense on Mutual Funds. But, he goes into a bit of detail on tactical asset allocation. And yes, you have captured the essence of what he said: valuations-based investing should not be part of the curriculum of Bogleheads 101. But if you have mastered of the basics in Bogleheads 101, TAA might be used sparingly with the cautions Jack made.
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

cjking
Posts: 1730
Joined: Mon Jun 30, 2008 4:30 am

Post by cjking » Tue Feb 01, 2011 6:10 am

wade wrote:1. A definition of PE10 is buried in there, but Graham and Dodd are also suggesting that looking at only PE10 is not enough, as interest rates should be in there somehow too. It's a good point that has been brought up earlier by peter71, and eventually I should update the regression model I used in a previous paper with (100/PE10), dividend yield, and interest rates in order to do this properly.


I think once you read "Valuing Wall Street" you may change your mind about using dividend yields or interest rates, assuming you find Smithers convincing. (I think you are covering ground he's been over, when looking for factors with predictive value.)

User avatar
BlueEars
Posts: 3510
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Post by BlueEars » Tue Feb 01, 2011 11:49 am


cjking
Posts: 1730
Joined: Mon Jun 30, 2008 4:30 am

Post by cjking » Tue Feb 01, 2011 3:57 pm

No, his earlier book.

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Tue Feb 01, 2011 8:59 pm

Thank you DRiP Guy, your comments mean a lot to me. Fulfilling my role as a professor, you might be expecting some kind of classical allusion. Though it is not obscure and pedantic enough to satisfy the tweed-clad pipe-smoking Platonic ideal of a professor, I think your strategy is much like that of Odysseus who had his men tie him to the mast of his ship so that he could not be tempted by the song of the sirens. Anyone who doesn't trust their ability to follow a mechanical rule no matter the circumstances would be best served taking the approach you suggest. With valuation-based investing, any behavioral mistakes will be amplified compared to always maintaining a fixed allocation, which is clearly not good. Buying a target date fund (that is based on underlying indexes and has low costs) and then forgetting about it is solid advice.

cjking: I think Mr. Smithers may have done himself a bit of a disservice by naming his books Valuing Wall Street and Wall Street Revalued, as it is kind of confusing. But I understand that his first book may be one of the very most important on this topic. I'm still waiting for my copy to arrive. I did get the 'halfway rule' programmed in now so I can start considering it as well.

lostcowboy
Posts: 212
Joined: Tue Sep 16, 2008 1:30 pm

Post by lostcowboy » Fri Feb 11, 2011 5:50 am

Hi Wade, I came across a sigma investing web page for you. Market Timing Strategies

User avatar
tadamsmar
Posts: 7361
Joined: Mon May 07, 2007 12:33 pm

Post by tadamsmar » Fri Feb 11, 2011 9:50 am

Wade:

All investors were suppose to pay cap gains taxes on market moves over most of the period of your study. Looks like you treated those taxes as if they were profits.

Of course, many investors can avoid cap gains taxes now. But, you would need to adopt the theory that the behavior of the markets are the same now as they were in the past (when selling off stock involved a big tax gain hits) in order to claim that a study that ignores taxes is relevant to today's situation. It seems more reasonable to assume that investors would adjust their behavior based on the tax implications causing the markets to behave differently. We have a limited amount of data on the more recent markets.

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 14, 2011 9:46 pm

Thank you lostcowboy and tadamsmar for the comments.

lostcowboy, I now have copies of both of the books by Lucile Tomlinson. I haven't had a chance to read them yet, but I'm looking forward to it. She sounds like a fascinating lady, but the only thing I've been able to determine regarding her biography is that she was from New York. I assume she must be deceased by now, or otherwise she is really old.

There is an interesting related article to this topic in the new February 2011 issue of the Journal of Financial Planning:

Is Buy and Hold Dead? Exploring the Costs of Tactical Reallocation
by David M. Blanchett
http://www.fpanet.org/journal/CurrentIs ... dHoldDead/

I read through it and he has an interesting take on the issue. First, he is saying that buy and hold is not dead in case that isn't clear. He believes fixed allocations can beat tactical asset allocations on a risk adjusted basis.

While I am trying to remain agnostic on the issue, I can already guess why supporters of long-term valuations-based tactical asset allocation will not accept this article as a final word on the issue. It is because he is setting a success rate for tactical asset allocation (such as you get things right 70% of the time), and then uses Monte Carlo simulations to compare a fixed allocation to a tactical allocation in which you are randomly assigned the best performing asset allocation based on the probability. Supporters of valuations-based TAA will not care that you get (for example) 70% correct calls on a random basis, but (for example) your 70% correct calls come at vital instances where the markets are out-of-whack and will provide the important wins.

Also, the paper assumes that a new asset allocation can be made each month, which is more frequent that the kind of stuff I am considering.

He does also make an effort to consider taxes. He still has to make some assumptions about when to call what a capital gain/loss or income, but his approach to taxes seems pretty sound. I hope that by the end of the year I can treat taxes in an even more formal way than this article, but right now I can't really say anything about taxes myself.

By the way, I've written a new paper which, depending on your interpretation, may or may not be relevant for this thread. The paper has a thread here:

viewtopic.php?t=68683

User avatar
tadamsmar
Posts: 7361
Joined: Mon May 07, 2007 12:33 pm

Post by tadamsmar » Tue Feb 15, 2011 12:10 pm

wade wrote:Thank you lostcowboy and tadamsmar for the comments.

lostcowboy, I now have copies of both of the books by Lucile Tomlinson. I haven't had a chance to read them yet, but I'm looking forward to it. She sounds like a fascinating lady, but the only thing I've been able to determine regarding her biography is that she was from New York. I assume she must be deceased by now, or otherwise she is really old.

There is an interesting related article to this topic in the new February 2011 issue of the Journal of Financial Planning:

Is Buy and Hold Dead? Exploring the Costs of Tactical Reallocation
by David M. Blanchett
http://www.fpanet.org/journal/CurrentIs ... dHoldDead/

I read through it and he has an interesting take on the issue. First, he is saying that buy and hold is not dead in case that isn't clear. He believes fixed allocations can beat tactical asset allocations on a risk adjusted basis.

While I am trying to remain agnostic on the issue, I can already guess why supporters of long-term valuations-based tactical asset allocation will not accept this article as a final word on the issue. It is because he is setting a success rate for tactical asset allocation (such as you get things right 70% of the time), and then uses Monte Carlo simulations to compare a fixed allocation to a tactical allocation in which you are randomly assigned the best performing asset allocation based on the probability. Supporters of valuations-based TAA will not care that you get (for example) 70% correct calls on a random basis, but (for example) your 70% correct calls come at vital instances where the markets are out-of-whack and will provide the important wins.

Also, the paper assumes that a new asset allocation can be made each month, which is more frequent that the kind of stuff I am considering.

He does also make an effort to consider taxes. He still has to make some assumptions about when to call what a capital gain/loss or income, but his approach to taxes seems pretty sound. I hope that by the end of the year I can treat taxes in an even more formal way than this article, but right now I can't really say anything about taxes myself.

By the way, I've written a new paper which, depending on your interpretation, may or may not be relevant for this thread. The paper has a thread here:

viewtopic.php?t=68683


My point was that if you did the backtesting while properly accounting for all historical tax rates, you would find that timing did not work. The paper you cited suggests that, but that's not really what he did.

I have seen plenty of papers with backtests that showed that timing works if you ignore the taxes, and then claim that you can use this in modern tax-deferred account. But I think it makes more sense to assume the future resembles the past. Timing did not work in the past with all costs accounted, therefore it won't work in the future with all costs accounted. That way everything makes sense, there is no need to explain why timers don't wipe out any inefficiencies because there are no apparent inefficiencies in the first place.

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Tue Feb 15, 2011 9:50 pm

tadamsmar wrote:My point was that if you did the backtesting while properly accounting for all historical tax rates, you would find that timing did not work. The paper you cited suggests that, but that's not really what he did.

I have seen plenty of papers with backtests that showed that timing works if you ignore the taxes, and then claim that you can use this in modern tax-deferred account. But I think it makes more sense to assume the future resembles the past. Timing did not work in the past with all costs accounted, therefore it won't work in the future with all costs accounted. That way everything makes sense, there is no need to explain why timers don't wipe out any inefficiencies because there are no apparent inefficiencies in the first place.


In relationship to what you are discussing, I guess an even bigger problem is that in the past no one could have even replicated the fixed allocation or valuations-based strategies because there were no index funds.

I'm not sure about your claim that using past tax rates, though, would completely overturn the results. But I am a bit ignorant of the history of U.S. tax law to be able to argue this forcefully. In the past, large institutional foundations such charitable foundations, pension funds, and university endowments were probably big market players. Were they taxed? And while Wikipedia suggests that capital gains taxes have been around for a long time, they were significantly increased in only the 1969 and 1976 tax laws. I'm not sure, but it seems like the income components of returns would have been taxed in the same way with either strategy, so it is really only the capital gains components we have to worry about. But again, I'm not suggesting answers here, only questions.

User avatar
BlueEars
Posts: 3510
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Post by BlueEars » Tue Feb 15, 2011 10:01 pm

The obvious way to avoid the "too many trades" argument and still do MT is to have very few trades/yr. I'd suggest trying for maybe 0.25 trades/yr on average i.e. 1 trade per 4 years. Definition of a trade is not the round trip but just 1 move.

There are many pretty decent arguments proposed to defeat MT ideas. Just read the 200 day moving average thread and you could pick them out quite easily.

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Thu Feb 24, 2011 9:08 pm

I'm working on a rebuttal to Fisher and Statman's 2006 paper, "Market Timing in Regressions and Reality".
http://onlinelibrary.wiley.com/doi/10.1 ... x/abstract

I'm not finished yet, but I have a figure that I think is interesting to share.

First, I don't agree that it is appropriate or fair to compare, especially on a risk-adjusted basis, a 100% stock strategy against a strategy that switches between 100% stocks and 100% bills based on whether PE10 is above or below its historical median. But in the first part of my paper, at least, I am playing by their rules.

One of the big criticisms of this kind of market timing is that you would have been sitting on the sidelines for the whole 1990s boom.

In this figure, the market timer was 100% Treasury bills for the whole time between the end of 1995 and the end of 2008, while the comparison non-market timer was 100% stocks.

The market timer looks pretty foolish by the end of 1999, and again by the end of 2006. But by the end of 2008, both guys were pretty much back together in terms of the total growth they experienced since the end of 1995.

Image

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy » Thu Feb 24, 2011 10:26 pm

Wade,

There is likely no way you can take my following comment except to think it comes from a position of false superiority, or hubris; but I don't think it does -- I feel it comes from empirical experience. My comment:

I wish there was a little handy look-up card that illustrated the phases of excitement, then disillusionment that newly minted market timing strategists are certain go through, because I am absolutely convinced these phases are distinct, discernible, definable, and predictable, if done by one skilled in the art.

I think you have entered a phase where, becoming cognizant of *other*, prior market timing schemes, one gleefully propels the proposed (and then refined and retooled, as often as required) market timing scheme of the author against those past attempts, thus building even more confidence that one is on the right path and that surely provable everlasting victory lies just ahead...

Godspeed Wade. I am certain your quest is in earnest, but I fear your result will be much as those who walked before you.

cjking
Posts: 1730
Joined: Mon Jun 30, 2008 4:30 am

Post by cjking » Fri Feb 25, 2011 10:37 am

In the real world, there are more alternatives than just being in or out of equities. For example, TIPS offered significantly higher prospective returns than equities at some point in the late nineties. From memory, the yield on property was much higher than the prospective yield on equities in many of the years around 2000 that a simple two-asset-class timer would have been out of equities.

There may be insufficient data to illustrate strategies that use other assets, but that doesn't mean there isn't good reason to implement such strategies.

wearethefall
Posts: 248
Joined: Sun Jul 29, 2007 6:01 am

Post by wearethefall » Sun Feb 27, 2011 5:00 pm

I'm reading Bogle's Little Book and he has some interesting things to say. He breaks total return into "investment return" and "speculative return". He says:

every decade of significantly negative speculative return was immediately followed by a decade in which it turned positive by a correlative amount -- the quiet 1910s and then the roaring 1920s, the dispiriting 1940s and then the booming 1950, the discouraging 1970s and then the soaring 1980s -- reversion to the mean writ large. ... Then, amazingly, there is an unprecedented second consectuive exuberant increase in speculative return in the 1990s, a pattern never before in evidence.


I guess recent out-of-sample data fits the narrative (although this book was written post-dot.com bursting) :D

Treadmill
Posts: 4
Joined: Sun Feb 27, 2011 9:21 pm

Timing v. Sentiment

Post by Treadmill » Sun Feb 27, 2011 9:44 pm

It's one thing to say timing doesn't work, but it's quite another to shut off one's emotions. I'd love to see anti-timer's sentiment charted against timer's sentiment. I bet it's pretty similar.

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Re: Timing v. Sentiment

Post by DRiP Guy » Sun Feb 27, 2011 11:00 pm

Treadmill wrote:It's one thing to say timing doesn't work, but it's quite another to shut off one's emotions. I'd love to see anti-timer's sentiment charted against timer's sentiment. I bet it's pretty similar.


Ah, but IMHO sentiment alone is unlikely to wreck a retirement plan!

I don't think we can say that about failed market timing.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 12:03 am

This is 2011, so data thru 2010 is certainly readily available. Therefore, the obvious question is why you stopped at 2008 instead of running through 2010. I suspect that the outcome would be remarkably different in favor of the buy-and-holder had you done so.

What am I missing?
Best Regards - Mel | | Semper Fi

DP
Posts: 659
Joined: Thu Apr 17, 2008 5:19 pm

Post by DP » Mon Feb 28, 2011 12:47 am

Mel Lindauer wrote:This is 2011, so data thru 2010 is certainly readily available. Therefore, the obvious question is why you stopped at 2008 instead of running through 2010. I suspect that the outcome would be remarkably different in favor of the buy-and-holder had you done so.

What am I missing?


I assume you are replying to the chart a few posts above that ends in 2008. It appears to me that what you are missing is that the timing model had far lower risk than the market, and so risk adjusted returns were much higher than buy and hold. I doubt that would change substantially once updated thru 2010, though I expect you are right that the total returns for buy and hold would be higher.

Don

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 1:08 am

Mel Lindauer wrote:This is 2011, so data thru 2010 is certainly readily available. Therefore, the obvious question is why you stopped at 2008 instead of running through 2010. I suspect that the outcome would be remarkably different in favor of the buy-and-holder had you done so.

What am I missing?


Mel, thank you for the comment. If you think I am trying to be sneaky, I think you are missing something important.

In the Fisher and Statman framework, the market timer would have finally returned to 100% stocks in 2009 so that they would have had the same returns that year and would still be together at the end of 2009.

For 2010, I don't have the market returns data yet, but the PE10 decision rule would have sent the market timer back to Treasury bills. I know stocks outperformed Treasury bills in 2010, and this would hurt the market timer.

Anyway, I just thought that figure was interesting. I know it is not the full story. I hope to have the full story for you in another couple of weeks.

But DP is right that the market timer enjoys a far less risky strategy. I can't understand why Fisher and Statman glossed over this rather essential detail. They comment that the Sharpe ratio for the market timer is biased upwards because it is comparable to someone with a 50/50 asset allocation rather than a 100/0 allocation. But that suggests to me that we should compare the market timer to a 50/50 allocation instead of a 100/0 allocation. But even without adjusting for risk, the market timer produces more or less the same returns over the whole historical period. That segment of the horserace I showed you began in 1871, not 1995.

And it took the unprecendented 1990s boom for the buy-and-hold guy to catch back up to the market timer. For most of history, the market timer was ahead. Relying on a 100% stocks strategy throughout the 1990s boom doesn't sound very low-risk to me.

robertalpert
Posts: 444
Joined: Wed Aug 22, 2007 10:09 pm

Post by robertalpert » Mon Feb 28, 2011 10:04 am

Check out Debra Wier's book: Timing the market, how to profit with yield curves (2005)

She also operates a blog at

http://timingthemarket.blogspot.com/

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 3:12 pm

wade wrote:
Mel Lindauer wrote:This is 2011, so data thru 2010 is certainly readily available. Therefore, the obvious question is why you stopped at 2008 instead of running through 2010. I suspect that the outcome would be remarkably different in favor of the buy-and-holder had you done so.

What am I missing?


Mel, thank you for the comment. If you think I am trying to be sneaky, I think you are missing something important.

In the Fisher and Statman framework, the market timer would have finally returned to 100% stocks in 2009 so that they would have had the same returns that year and would still be together at the end of 2009.

For 2010, I don't have the market returns data yet, but the PE10 decision rule would have sent the market timer back to Treasury bills. I know stocks outperformed Treasury bills in 2010, and this would hurt the market timer.

Anyway, I just thought that figure was interesting. I know it is not the full story. I hope to have the full story for you in another couple of weeks.

But DP is right that the market timer enjoys a far less risky strategy. I can't understand why Fisher and Statman glossed over this rather essential detail. They comment that the Sharpe ratio for the market timer is biased upwards because it is comparable to someone with a 50/50 asset allocation rather than a 100/0 allocation. But that suggests to me that we should compare the market timer to a 50/50 allocation instead of a 100/0 allocation. But even without adjusting for risk, the market timer produces more or less the same returns over the whole historical period. That segment of the horserace I showed you began in 1871, not 1995.

And it took the unprecendented 1990s boom for the buy-and-hold guy to catch back up to the market timer. For most of history, the market timer was ahead. Relying on a 100% stocks strategy throughout the 1990s boom doesn't sound very low-risk to me.


Sure, Wade, but if you choose any other time period, the B&H far outpaced the market timers. So what's the point except that it appears to be cherry-picking a period that makes your point. Change the time period to 1999 or 2006 and you change the entire picture and the results are dramatically different. There's only one period in the entire sequence where your assettion is true, and that's the period you choose to use? Sure looks like data mining to me.
Best Regards - Mel | | Semper Fi

User avatar
tegdirb
Posts: 24
Joined: Tue Feb 15, 2011 3:00 pm
Location: Central PA

Post by tegdirb » Mon Feb 28, 2011 5:54 pm

Harry Dent has some long term strategy ideas in his program "How to prosper in a Downturn".

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 8:45 pm

Mel Lindauer wrote:Sure, Wade, but if you choose any other time period, the B&H far outpaced the market timers. So what's the point except that it appears to be cherry-picking a period that makes your point. Change the time period to 1999 or 2006 and you change the entire picture and the results are dramatically different. There's only one period in the entire sequence where your assettion is true, and that's the period you choose to use? Sure looks like data mining to me.


I take the issue of data mining very seriously, and with all due respect, any data mining that I am doing is in favor of buy-and-hold, not in favor of market timing.

I don't think either the 100% stocks buy-and-hold, or the market timing strategy which switches between 100% stocks and 100% Treasury bills, are at all realistic or feasible for a conservative long-term investor. But for now, let's play by the Fisher and Statman rules, as the findings for "market timing" are so robust anyway, that it hardly matters how we do it.

As is considered in Fisher and Statman, this table shows the results of a horserace between 100% stocks, and market timing over the entire period, January 1871 to January 2010, for $1 invested in 1871.

Image

100% stocks provides final wealth of $95,404.

I consider 4 different decision rules for market timing. The one used by Fisher and Statman, "historical median" provides final wealth of $124,147. But the rolling mean and rolling median approaches are more realistic. The worst performing one of these is the "rolling median" one, and with that the market timer has final wealth of $94,866, which is slightly below the buy-and-hold. In the example I gave above, I data-mined in buy-and-hold's favor by basing it on the market-timing strategy that gave the worst results for market timing.

I think a fair conclusion is that market timing will provide you with approximately the same wealth as 100% stocks buy-and-hold.

But before going any further, let's consider risk. 100% stocks is riskier by any imaginable criteria. Its standard deviation of portfolio returns is much higher at 18.02%, its Sharpe ratio defined in two different ways is lower than any market timing. Fisher and Statman argue that the Sharpe ratio is biased, so I also add the "Information Ratio" which compares market timing to a 50/50 fixed portfolio that has the same ex-ante average stock allocation. This ratio compares the active returns from market timing to the active risk of market timing, and the fact that these numbers are positive all help to suggest that market timing is providing superior risk-adjusted returns. Next, the maximum drawdown from market timing is much less. That is how far the portfolio drops from past highs to current lows. The buy-and-hold once experienced a 60.96% drop, whereas the worst drop market timing was 24.16%. Considering returns over maximum drawdown, market timing still does much better. Then, there is another set of risk measures which considers that investors are more sensitive to losses rather than gains. The downside deviation which only penalizes returns less than zero shows that market timing is much less risky, and this in turn favors market timing for the Sortino ratio. Market timing also has a much lower average stock allocation, and market timing changes its allocation on average once every 5 years.

So, market timing provides essentially the same returns, but with much less risk.

This is in spite the fact that market timing has to overcome one major obstacle in its battle with 100% stocks buy-and-hold: the very generous average equity premium enjoyed by U.S. stock investors in the past.

I think it is more realistic to compare market timing to a fixed 50/50 allocation rebalanced each year. With that, the wealth accumulation is only $13,426. That's the equity premium in a nut shell. Fixed 50/50 provides much less wealth than fixed 100/0. At least in historical U.S. data. Mr. Bogle would suggest not to count on this necessarily continuing in the future.

Ex-ante, I would have expected market timing to be much riskier than fixed 50/50. That is because the market timing strategy is rather extreme and unrealistic anyway, as I explained at the start. But comparing risk measures here, market timing is holding its own. It does have higher standard deviations, higher downside deviations and lower Sortino ratios. At the same time, it has higher Sharpe ratios, and I already discussed its positive information ratios which are compared to 50/50, and it has lower maximum drawdowns, and higher returns over maximum drawdowns. Compared to 50/50, I think it is fair to conclude that market timing provides signficantly higher returns at a comparable level of risk.

Image

Now, more specifically, to the point you are making about the choice of end period, and how I am data mining here. I needed to bring up the table above to show how I already am helping out buy-and-hold by comparing it to the worst-case for market timing. This new figure shows the horserace over the entire historical period. Looking at each possible date as the end period:

-market timing provides more wealth in 51.8% of cases
-buy-and-hold provides more wealth in 32.37% of cases
-they are essentially tied in 15.83% of cases, where I define being tied as the difference in wealth accumulations between the two strategies as being less than 1% of the value of the buy-and-hold wealth accumulation. I think there is no point in declaring a winner in cases where they are so close.

So, if you wish to focus on one of the 32.37% of cases in which buy-and-hold is clearly ahead, I respectfully submit that you are the one doing data mining. Even though market-timing is ahead more often than not, I'd be willing to call it a draw on this point, and instead focus on how market timing provides substantially less risk for the same general returns.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 9:21 pm

Hi Again Wade:

Are you accounting for the costs of the market timing scheme (ongoing annual taxes on the interest and ST and LTCG on the equtiies, which reduce the amount available for reinvesting in the MT scheme)? If not, then it's not a valid comparison with a tax-efficient B&H equity portfolio.

It appears you're charting the full amount of the market timer's porfolio each time you make a MT switch despite the fact that the investor would be paying taxes on an ongoing basis with each switch, thus reducing the amount that's available to reinvest in each of the switches.

Do I understand your methodology correctly and if so, it would appear that you need to make adjustments for those ongoing MT expenses.

Regards,

Mel
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 9:32 pm

Mel Lindauer wrote:Hi Again Wade:

Are you accounting for the costs of the market timing scheme (ongoing annual taxes on the interest and ST and LTCG on the equtiies, which reduce the amount available for reinvesting in the MT scheme)? If not, then it's not a valid comparison with a tax-efficient B&H equity portfolio.

It appears you're charting the full amount of the market timer's porfolio each time you make a MT switch despite the fact that the investor would be paying taxes on an ongoing basis with each switch, thus reducing the amount that's available to reinvest in each of the switches.

Do I understand your methodology correctly and if so, it would appear that you need to make adjustments for those ongoing MT expenses.

Regards,

Mel


I didn't make adjustments for taxes. We've been through this issue already in this thread. If you don't like me to say: just put it in a tax-deferred vehicle, or in a Roth IRA, then whose taxes should I consider? ST capital gains taxes might be an issue for this extreme market timing example, but for a more reasonable market timing case, only long-term capital gains taxes would apply. And rates for these just increased to the relative levels we know today in the 1960s and 1970s. Before then, it wasn't such an issue. Should I apply past rates or current rates? I'm not going to do this now, but I would like to do it carefully by the end of the year.

I acknowledge that taxes will eat into some of the market timer's returns.

At the same time, it is hard to imagine an even remote possible that the taxes would be high enough to overturn everything I just finished describing.

DP
Posts: 659
Joined: Thu Apr 17, 2008 5:19 pm

Post by DP » Mon Feb 28, 2011 9:38 pm

Hi,
The taxes get complicated on the buy and hold side as well. I don't think anyone is going to hold stocks for some 140 years and not owe some taxes along the way.

Don

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 9:56 pm

DP wrote:Hi,
The taxes get complicated on the buy and hold side as well. I don't think anyone is going to hold stocks for some 140 years and not owe some taxes along the way.

Don


1. The point is that taxes on the market-timing scheme have to be paid on an annual basis and at the highest tax rates when one goes to cash.
2. As Wade acknowledged, there may also be ST gains taxes to pay for the market-timer, again at the highest personal tax rate.
3. Both #1 and #2 reduce the amount that's available for reinvesting in the market-timing scheme, but they're not account for in Wade's charts or his study.
4. On the other hand, a tax-efficient equity portfolio would only owe the lower LTCG tax rate when sold. However, if left to an heir, they'd get the stepped-up cost basis and NO tax would be due.
Best Regards - Mel | | Semper Fi

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 10:07 pm

wade wrote:
Mel Lindauer wrote:Hi Again Wade:

Are you accounting for the costs of the market timing scheme (ongoing annual taxes on the interest and ST and LTCG on the equtiies, which reduce the amount available for reinvesting in the MT scheme)? If not, then it's not a valid comparison with a tax-efficient B&H equity portfolio.

It appears you're charting the full amount of the market timer's porfolio each time you make a MT switch despite the fact that the investor would be paying taxes on an ongoing basis with each switch, thus reducing the amount that's available to reinvest in each of the switches.

Do I understand your methodology correctly and if so, it would appear that you need to make adjustments for those ongoing MT expenses.

Regards,

Mel


I didn't make adjustments for taxes. We've been through this issue already in this thread. If you don't like me to say: just put it in a tax-deferred vehicle, or in a Roth IRA, then whose taxes should I consider? ST capital gains taxes might be an issue for this extreme market timing example, but for a more reasonable market timing case, only long-term capital gains taxes would apply. And rates for these just increased to the relative levels we know today in the 1960s and 1970s. Before then, it wasn't such an issue. Should I apply past rates or current rates? I'm not going to do this now, but I would like to do it carefully by the end of the year.

I acknowledge that taxes will eat into some of the market timer's returns.

At the same time, it is hard to imagine an even remote possible that the taxes would be high enough to overturn everything I just finished describing.


Hi Again Wade:

Problem with your "put it in an IRA" is that Bogleheads understand that one should ideally hold tax-efficient equities in their taxable account. Otherwise, they're converting what could be LTCG taxes to ordinary income. And Roths do have income limits, so that's not always possible. And, for folks who want to take the tax-deduction, the IRA or their 401k plan make more sense than a Roth.

So ideally, they'd hold tax-efficient equities in their taxable account and bonds and other tax-inefficient things in their tax-deferred accounts.

Finally, it's possible for the investor to escape taxes totally on the tax-efficient taxable account when left to their heirs.
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 10:15 pm

Mel Lindauer wrote:1. The point is that taxes on the market-timing scheme have to be paid on an annual basis and at the highest tax rates when one goes to cash.
2. As Wade acknowledged, there may also be ST gains taxes to pay for the market-timer, again at the highest personal tax rate.
3. Both #1 and #2 reduce the amount that's available for reinvesting in the market-timing scheme, but they're not account for in Wade's charts or his study.
4. On the other hand, a tax-efficient equity portfolio would only owe the lower LTCG tax rate when sold. However, if left to an heir, they'd get the stepped-up cost basis and NO tax would be due.


1. Only when done in a taxable account.

2. Only for this extreme market timing. Actually, less extreme market timing, such as simply switching to 50% stocks instead of 0% stocks when the market is overvalued, results in even more wealth and can surely be done with only paying long-term capital gains taxes.

3. So it just comes down to this: you are going to ignore everything I wrote above and just base your entire case now on this nebulous tax issue? That's fine for you, but it's frustrating. I feel like there is not much point in investigating taxes further just to satisfy you, because if I do that you will just jump to the next thing on your laundry list. What happen to your accusations about data mining?

4. If no one is ever planning to spend any if the savings, then what is the point?

--

Edit: I didn't see your second post before. I agree that my IRA argument is not fully effective in this case.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 10:47 pm

wade wrote:
Mel Lindauer wrote:1. The point is that taxes on the market-timing scheme have to be paid on an annual basis and at the highest tax rates when one goes to cash.
2. As Wade acknowledged, there may also be ST gains taxes to pay for the market-timer, again at the highest personal tax rate.
3. Both #1 and #2 reduce the amount that's available for reinvesting in the market-timing scheme, but they're not account for in Wade's charts or his study.
4. On the other hand, a tax-efficient equity portfolio would only owe the lower LTCG tax rate when sold. However, if left to an heir, they'd get the stepped-up cost basis and NO tax would be due.


1. Only when done in a taxable account.

2. Only for this extreme market timing. Actually, less extreme market timing, such as simply switching to 50% stocks instead of 0% stocks when the market is overvalued, results in even more wealth and can surely be done with only paying long-term capital gains taxes.

3. So it just comes down to this: you are going to ignore everything I wrote above and just base your entire case now on this nebulous tax issue? That's fine for you, but it's frustrating. I feel like there is not much point in investigating taxes further just to satisfy you, because if I do that you will just jump to the next thing on your laundry list. What happen to your accusations about data mining?

4. If no one is ever planning to spend any if the savings, then what is the point?

--

Edit: I didn't see your second post before. I agree that my IRA argument is not fully effective in this case.


Hi Again Wade:

I'm just trying to get to the bottom of all the issues, before you attempt to publish what may be found to be inferior or incomplete work by your peer reviewers.

You appear to honestly want to pick our brains, so I'm giving you the feedback you asked for. These are issues that will be raised later, so you need to face them now. However, it appears you're getting just a little bit testy because I'm raising issues that you may not have considered and which could certainly change the results.

Better to get the issues raised here and resolved instead of having your paper rejected because of these very basic issues which you haven't addressed.
Last edited by Mel Lindauer on Mon Feb 28, 2011 10:55 pm, edited 1 time in total.
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 10:55 pm

Mel Lindauer wrote:I'm just trying to get to the bottom of all the issues, Wade. You appear to honestly want to pick our brains, so I'm giving you the feedback you asked for. However, it now it appears you're getting a bit testy because I'm raising issues that you may not have not have considered which skews the results.


I'll try not to get testy.

When we stop comparing apples and oranges, and instead compare two scenarios which offer broadly similar risks to investors:

* the worst performing market timing scenario produced $94,866

* a 50/50 asset allocation produced $13,426

the 50/50 asset allocation is also going to have to pay some capital gains taxes when it rebalances annually, but the market timer will surely have to pay more taxes over time.

But do you honestly believe that the extra taxes paid by the market timer will wipe out the entire surplus gains it had provided at a broadly similar level of risk?

-----

Thank you for the comments at any rate. Earlier, a commenter suggest that I should apply historical tax rates. It sounds like you are suggesting that I apply current tax rates. Is this correct? Thank you.
Last edited by wade on Mon Feb 28, 2011 10:58 pm, edited 1 time in total.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 10:57 pm

wade wrote:
Mel Lindauer wrote:I'm just trying to get to the bottom of all the issues, Wade. You appear to honestly want to pick our brains, so I'm giving you the feedback you asked for. However, it now it appears you're getting a bit testy because I'm raising issues that you may not have not have considered which skews the results.


I'll try not to get testy.

When we stop comparing apples and oranges, and instead compare two scenarios which offer broadly similar risks to investors:

* the worst performing market timing scenario produced $94,866

* a 50/50 asset allocation produced $13,426

the 50/50 asset allocation is also going to have to pay some capital gains taxes when it rebalances annually, but the market timer will surely have to pay more taxes over time.

But do you honestly believe that the extra taxes paid by the market timer will wipe out the entire surplus gains it had provided at a broadly similar level of risk?


The figures in your original chart were 100%/0 and now you're talking about the returns for a 50/50 portfolio?
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 11:00 pm

Mel Lindauer wrote:
wade wrote:
Mel Lindauer wrote:I'm just trying to get to the bottom of all the issues, Wade. You appear to honestly want to pick our brains, so I'm giving you the feedback you asked for. However, it now it appears you're getting a bit testy because I'm raising issues that you may not have not have considered which skews the results.


I'll try not to get testy.

When we stop comparing apples and oranges, and instead compare two scenarios which offer broadly similar risks to investors:

* the worst performing market timing scenario produced $94,866

* a 50/50 asset allocation produced $13,426

the 50/50 asset allocation is also going to have to pay some capital gains taxes when it rebalances annually, but the market timer will surely have to pay more taxes over time.

But do you honestly believe that the extra taxes paid by the market timer will wipe out the entire surplus gains it had provided at a broadly similar level of risk?


The study was 100%/0 and now you're talking about 50/50. Which is it?


Please refer to my post above that includes a table and a figure. I talk about both cases there. I'm not switching anything.

Would you agree that it is not appropriate to compare the returns of different strategies that result from taking very different levels of risk, and that we should compare risk-adjusted returns? In that case, it is 50/50.

Am I the only one who thinks that it isn't quite right to be comparing a strategy that is 100% stocks against a strategy that is on average only 50% stocks? Fisher and Statman do this, but one of the points I am trying to make is that this isn't a fair comparison. Am I wrong?

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 11:06 pm

wade wrote:
Mel Lindauer wrote:
wade wrote:
Mel Lindauer wrote:I'm just trying to get to the bottom of all the issues, Wade. You appear to honestly want to pick our brains, so I'm giving you the feedback you asked for. However, it now it appears you're getting a bit testy because I'm raising issues that you may not have not have considered which skews the results.


I'll try not to get testy.

When we stop comparing apples and oranges, and instead compare two scenarios which offer broadly similar risks to investors:

* the worst performing market timing scenario produced $94,866

* a 50/50 asset allocation produced $13,426

the 50/50 asset allocation is also going to have to pay some capital gains taxes when it rebalances annually, but the market timer will surely have to pay more taxes over time.

But do you honestly believe that the extra taxes paid by the market timer will wipe out the entire surplus gains it had provided at a broadly similar level of risk?


The study was 100%/0 and now you're talking about 50/50. Which is it?


Please refer to my post above that includes a table and a figure. I talk about both cases there. I'm not switching anything.

Would you agree that it is not appropriate to compare the returns of different strategies that result from taking very different levels of risk, and that we should compare risk-adjusted returns? In that case, it is 50/50.


It's a very personal thing. Each investor determines the amount of risk they're willing to take, and many are willing to take on more risk than others for the expected higher returns. Some investors choose 100% equity, some might be 80/20, some 60/40 etc. So you can't say that all investors should or do wish to assume the same level of risk.
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 11:16 pm

Mel Lindauer wrote: It's a very personal thing. Each investor determines the amount of risk they're willing to take, and many are willing to take on more risk than others for the expected higher returns. Some investors choose 100% equity, some might be 80/20, some 60/40 etc. So you can't say that all investors should or do wish to assume the same level of risk.


I'm definitely not saying that all investors should take on the same risk. Of course not.

What I am saying, or asking, is that if you wish to compare two different investment strategies to see which can be expected to provide the highest returns, shouldn't you make some kind of adjustment for risk? If you ignore risk, then whatever provides the highest expected return over time will tend to win. Is it okay to ignore risk for the comparisons? Or should you only try to compare strategies which offer the same risks? Then, from there, you can move on and make further adjustments to the strategy to calibrate the level of risk you are willing to incur as an individual investor.

To give some context. I am currently 70/30. But I am seriously considering changing to 50% stocks/70% stocks/90% stocks depending on the level of PE10. I won't do this until at least the summer, because I need to start studying for the CFA Level 3 exam soon, and I can see that it will teach lots of important tools for making these comparisons, and give me more time to think about this rather serious and important decision. I still have 35 or 40 years before I would seriously consider retiring.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Mon Feb 28, 2011 11:36 pm

wade wrote:
Mel Lindauer wrote: It's a very personal thing. Each investor determines the amount of risk they're willing to take, and many are willing to take on more risk than others for the expected higher returns. Some investors choose 100% equity, some might be 80/20, some 60/40 etc. So you can't say that all investors should or do wish to assume the same level of risk.


I'm definitely not saying that all investors should take on the same risk. Of course not.

What I am saying, or asking, is that if you wish to compare two different investment strategies to see which can be expected to provide the highest returns, shouldn't you make some kind of adjustment for risk? If you ignore risk, then whatever provides the highest expected return over time will tend to win. Is it okay to ignore risk for the comparisons? Or should you only try to compare strategies which offer the same risks? Then, from there, you can move on and make further adjustments to the strategy to calibrate the level of risk you are willing to incur as an individual investor.

To give some context. I am currently 70/30. But I am seriously considering changing to 50% stocks/70% stocks/90% stocks depending on the level of PE10. I won't do this until at least the summer, because I need to start studying for the CFA Level 3 exam soon, and I can see that it will teach lots of important tools for making these comparisons, and give me more time to think about this rather serious and important decision. I still have 35 or 40 years before I would seriously consider retiring.


I'm simply raising issues that others are certain to raise. And ignoring the cost of taxes in the market-timing strategy is a major flaw IMO.

Do you have clearly-defined and executable buy and sell points where all of the information needed to follow the market-timing system is publicly available to all investors?

Finally, have you considered what the market impact would be if a very large number of investors followed the system and executed their market timing buys and sells on the same day? (I'd guess that there would only be one right buy and sell date for the data to be valid).

PS - Congrats on getting past the first two CFA tests. You're almost there!
Best Regards - Mel | | Semper Fi

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Mon Feb 28, 2011 11:46 pm

Mel Lindauer wrote:I'm simply raising issues that others are certain to raise. And ignoring the cost of taxes in the market-timing strategy is a major flaw IMO.

Do you have clearly-defined and executable buy and sell points where all of the information needed to follow the market-timing system is publicly available to all investors?

Finally, have you considered what the market impact would be if a very large number of investors followed the system and executed their market timing buys and sells on the same day? (I'd guess that there would only be one right buy and sell date for the data to be valid).

PS - Congrats on getting past the first two CFA tests. You're almost there!


Thank you for these comments. I think we can wrap up our discussion, thank you very much.

- Taxes do remain an area I haven't explored, but would like to. But it will require a lot of work so I cannot do it until after the exam.

-About the decision rules with publicly available information. That should not be a problem. After the market closes at the end of the year, people should be able to get an update on PE10 immediately and change their asset allocation for the following year when necessary.

-About the potential market impacts of many people adopting this strategy, I don't have any clear answer. Some may think it will eliminate booms and busts from the market, but as a "dismal scientist" (a nickname for economists) I can't accept that. There are always unintended consequences to new policies. Bubbles would just form elsewhere. This is an important question though. Under my assumptions, people make their moves on January 1st of each year, but I suppose in reality things wouldn't work out quite that precisely. But at the end of the day, as well, it will probably be hard to convince a lot of people to overcome the psychological roadblocks of this contrarian investment strategy, so that the market impacts may not be so large.

After all, as you kindly pointed out much earlier, all of this has been around at least since the time of Graham and Dodd.

Thank you again for all your comments and time today, Wade

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Sat Mar 12, 2011 12:05 am

Hi, my family and I made it through the earthquake. Tokyo had quite a shake, but all of the tragedies are occurring a couple hundred miles up the coast where the tsunami hit. I was in my office when the earthquake hit and it was scary as all my bookshelves and file cabinets emptied around me. My office is on the 9th floor of a building specifically designed to sway a lot with earthquakes. Our apartment was hardly affected at all.

Anyway, though it is a bit anticlimatic, as I discussed lots of the results here with Mel last week, I did finish my paper about the Fisher and Statman study on market timing, if anyone is interested to see the final version. I did incorporate some utility-based risk measures that I think are more persuasive to show that the valuation-based timing approach was less risky:

"Revisiting the Fisher and Statman Study on Market Timing"
http://ideas.repec.org/p/pra/mprapa/29448.html

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy » Sat Mar 12, 2011 8:34 am

Watching the news reports, it just gets more and more shocking how devastating the quake was, and how widespread it's aftermath continues to be.

Best wishes to you, and all in the affected area, Wade.

User avatar
Mel Lindauer
Moderator
Posts: 27373
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Post by Mel Lindauer » Sat Mar 12, 2011 1:09 pm

Glad to hear that you're OK, Wade.
Best Regards - Mel | | Semper Fi

User avatar
Pres
Posts: 230
Joined: Thu Aug 07, 2008 4:25 pm
Location: Eurozone

Post by Pres » Sun Mar 13, 2011 7:15 pm

wade wrote:Hi, my family and I made it through the earthquake. Tokyo had quite a shake, but all of the tragedies are occurring a couple hundred miles up the coast where the tsunami hit. I was in my office when the earthquake hit and it was scary as all my bookshelves and file cabinets emptied around me. My office is on the 9th floor of a building specifically designed to sway a lot with earthquakes. Our apartment was hardly affected at all.

Anyway, though it is a bit anticlimatic, as I discussed lots of the results here with Mel last week, I did finish my paper about the Fisher and Statman study on market timing, if anyone is interested to see the final version. I did incorporate some utility-based risk measures that I think are more persuasive to show that the valuation-based timing approach was less risky:

"Revisiting the Fisher and Statman Study on Market Timing"
http://ideas.repec.org/p/pra/mprapa/29448.html


Glad you're allright, Wade.

Interesting paper!

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Wed Mar 16, 2011 10:58 pm

Thank you for the warm wishes.

My family and I will be heading to Iowa tomorrow for a few weeks. Although I think we are overreacting by leaving, we do have a 3-year old son and my wife is 3-months pregnant. The radiation that reached Tokyo on Tuesday morning (though it was still low enough to be safe) made me a bit more concerned than I had been up to that point.

With all this going on, I came to the realization that I should stop trying to finish my articles on this topic until after the CFA exam in June.

So that I could help myself to just forget about this until June, I did write up two blog entries with some preliminary results. It is just some food for thought. I hope to finish these in June, as I said, and one of the big things that has been slowing my progress was to be able to write a very detailed "caveats" section.

Here are links for my two blog posts:

"Long-Term Conservative Investors and Valuation-Based Asset Allocation Strategies"
http://wpfau.blogspot.com/2011/03/long- ... s-and.html

"Could Valuation-Based Asset Allocation Strategies Have Worked in Japan?"
http://wpfau.blogspot.com/2011/03/could ... ation.html

User avatar
fredflinstone
Posts: 2069
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone » Sat May 28, 2011 9:56 pm

I read the paper today at http://mpra.ub.uni-muenchen.de/29448/1/ ... _29448.pdf and I must say it is outstanding. I believe your paper deals a serious blow to the standard Boglehead notion that market-timing can't work. Thank you also for presenting your methods and results so clearly, and for writing in a language that a layman can understand.

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Post by wade » Sat May 28, 2011 10:37 pm

Thank you Fred.

Market timing usually has a short-term focus, whereas what I am talking about requires a rather long-term horizon. I think this can co-exist with the Boglehead's philosphy for the most part.

The paper you read was really meant as just a secondary study as it considers just the full 1871-2010 period together, and as it only deals with rather extreme strategies of all-stocks or all-bonds depending on whether PE10 is above or below its average, by however much.

I'm working on another paper which considers shorter rolling periods and more realistic asset allocation stategies, and I hope it can be finished in another month or two. This other (unfinished) paper will be more relevant for individual investors.

Of course, there are all kinds of interesting caveats and behavioral issues to consider for this type of approach. Trying to write those all out in a clear and concise way is what was mainly holding me back from finishing the other paper so far.

Best wishes, Wade

Post Reply