Momentum based investing in real world

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Momentum based investing in real world

Post by larryswedroe » Fri Apr 19, 2013 10:15 am

Thought this would be of interest, look at live results for AQR's 3 funds with 3 years live data
Note I will be now contributing at IU site about once a week

Best wishes
Larry

User avatar
NoRoboGuy
Posts: 732
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: MOMENTUM based investing in real world

Post by NoRoboGuy » Fri Apr 19, 2013 10:34 am

So this could be used as a tilt, just like one might tilt to value? To what would you attribute the disparity in tracking error between the small and large caps?
There is no free lunch.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: MOMENTUM based investing in real world

Post by larryswedroe » Fri Apr 19, 2013 11:36 am

Louis
Any well run fund that gives up trying to replicate a benchmark to get more effective trading is going to have large tracking error. That's something you have to learn to live with if you use funds like DFA or Bridgeway. They and their investors accept the RANDOM tracking error to improve expected returns. What you have to look for is are the tracking errors truly random. For that you have to do what is called an attribution analysis, which the retail investor is never going to get, or not likely. It's one of the reasons why funds like DFA don't work with public. Trying to explain that issue and having investors understand and accept it is not easy and very time consuming

I hope that is helpful
Larry

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: MOMENTUM based investing in real world

Post by EDN » Fri Apr 19, 2013 12:37 pm

louis c wrote:So this could be used as a tilt, just like one might tilt to value? To what would you attribute the disparity in tracking error between the small and large caps?
If you hold a value fund, or even large and small value, a momentum fund is not what I would use to counterbalance it. Let's set aside the fact that momentum is an anomaly (which means it could persist, or could not), and your momentum fund must be run by a firm that can absorb the costs of about 100% turnover through patient trading (not easy). Let's evaluate the claim from momentum proponents:

They say you should buy a basket of stocks that has recently just gone up quite a bit with the assumption that it will continue to go up for a little while longer...and you'll be able to sell the appreciated securities prior to a major reversal. That's it -- offset stocks with low prices that you expect to reverse over a matter of years (i.e. value stocks) with stocks that have gone up quite a bit recently and should continue to do so for a matter of months (i.e. high momentum stocks). It's up to you to evaluate if that is the best way to deploy your wealth.

I'll just mention, there is an alternative. We could just stick with basic finance:

Value stocks have higher expected returns because they have low prices for a given level of fundamentals (book value, earnings, etc.).

So we got the low price thing covered. What is the best way to counterbalance low-priced stocks? With high-priced stocks, of course. And that is what we've traditionally done, pairing Value with a Growth Index (lowest returns) or the S&P 500 or other market index (higher than pure growth returns, but not as much of an offsetting effect). We just had to choose which was more important -- less tracking error (we'd choose growth) or higher returns and greater tracking error (we'd choose the S&P 500).

But "price" is only one side of the expected return equation (the denominator, to be exact):

Expected Returns = Expected Cash Flows / Price

With this model in mind, you could also target expected cash flows (the numerator), as that is the other side of expected returns. Companies with the highest future expected cash flows (which tend to be the companies who are the most profitable today) also have high expected returns.

When you pair high-priced stocks (that counterbalance value) that have high current profitability, you accomplish your objective much easier and more predictably:

(1) adding large growth stocks to value reduces the tracking error of your portfolio and increases diversification. When "value" stumbles, it is expected that high profit growth stocks should do the best (and vice versa).
(2) focusing on the most profitable growth companies means you aren't stuck with the lowest expected returns we traditionally see from growth indexes that just look to hold high P/E or P/B stocks with no consideration of profitability. For example, from 1979-2012, the Russell 1000 Growth earned +10.6%, the Russell 1000 Value earned +12.0%. A large growth index with an emphasis on the most highly profitable stocks earned +11.9%.

Relative to the traditional approach of using a Growth Index or the S&P 500, this way will (a) be just as efficient at reducing value tracking error, and (b) have even higher expected returns than using the S&P 500. That is a good tradeoff.

I can't tell you the tickers or if there are large growth funds or ETFs that specifically target highly profitable companies right now, I don't follow the ETF world as closely as some, but there will be, and those funds (look for low costs, high diversification, and reputable management) will be the best compliment to a traditional value portfolio.

Eric

EDN
Posts: 528
Joined: Sat Dec 29, 2012 7:55 pm
Location: Oklahoma City, OK

Re: MOMENTUM based investing in real world

Post by EDN » Fri Apr 19, 2013 12:49 pm

FWIW, I view the approach that unbundles (for the most part) low-priced value from high-priced profitability as a better one for investors than one that tries to do everything (profitability, value, and momentum) all inside a core fund. There maybe some trading efficiencies there (relative to a pure long-only momentum fund), but you loose the visual exposure to the extreme offsetting results.

From 1995-1999, the S&P 500 did 28% per year but Russell 1000 Value only did 23%. It'd been nice to see the separate Large Growth Index earning its 31% to give you some indication you were diversified and exposed to that side of the market.

From 2007-2008, we had one year (2007) where the market was up modestly but value got killed, and then the next year everything lost a lot. During these two years, the Russell 1000 Value lost -37%, Large Growth only lost -23% (the S&P 500 lost -33%).

In 2011, value took a sharp turn for the worse in the Summer/Fall. For the year, like 2007, the S&P 500 was up (2%), value was negative, but large growth was up over 6% for the year.

Also, each investor should be deciding how much exposure they want to growth (profitability) vs. value (distress). The Google employee may want to go pretty light on growth, say 10% growth (profitability), 30% large value, 60% small value. The guy that runs the Ford dealership, maybe something more like 50% growth, 30% large value, and 20% small value. You just cannot do that in a commingled fund.

Eric

User avatar
nisiprius
Advisory Board
Posts: 36864
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Momentum based investing in real world

Post by nisiprius » Fri Apr 19, 2013 4:27 pm

If we are going to consider a three-year period as indicative, then, in the real world, since inception of the AQR funds, $10,000 grew to

--$610.67 more in Vanguard Large-Cap Index (VLACX) than it did in AQR Momentum (AMOMX)
--$1705.23 more in Vanguard Small-Cap Index Fund (NAESX) than in AQR Small-Cap Momentum (ASMOX).
--$886.06 more in AQR International Momentum than Vanguard Total International Index Fund (VGTSX).

I am not sure what would be the most appropriate way to strike a total overall score, but I would regard that a mixed and inconclusive result.

But perhaps a three-year period isn't actually indicative of anything.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Momentum based investing in real world

Post by larryswedroe » Fri Apr 19, 2013 4:31 pm

nisprius
Don't think that is good idea, comparing short term returns of factors not very useful
The purpose of the piece was to show whether or not the strategy was IMPLEMENTABLE or the whether the frictions were so high that the theoretical returns were just that, and could not be earned in real world
It's like the issue of the January effect-was there only in theory as all the gains were in the tiniest of stocks, not just small stocks, where the spreads equaled the premium, not even counting market impact costs
Larry

asset_chaos
Posts: 1388
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: Momentum based investing in real world

Post by asset_chaos » Fri Apr 19, 2013 5:32 pm

Are the momentum fund returns pre- or post-tax? I imagine momentum stratgies have high turnover---and M* says around 70% for the AQR funds, which truth be told is not as high as I'd have guessed. Nonetheless, that's still pretty high. What's that do to taxable distributions for the fund investors?

I still have no understanding of what the risk is that should generate a momentum return, while cheerfully stipulating that the data is unambiguous that momentum exists.
Regards, | | Guy

User avatar
nisiprius
Advisory Board
Posts: 36864
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Momentum based investing in real world

Post by nisiprius » Fri Apr 19, 2013 6:03 pm

asset_chaos wrote:I still have no understanding of what the risk is that should generate a momentum return, while cheerfully stipulating that the data is unambiguous that momentum exists.
I still have no understanding of why an effect that has been known for seventy-six years would persist. The 1937 paper is:

Cowles 3rd, Alfred E. and Herbert E. Jones (1937). "Some A Posteriori Probabilities in Stock Market Action". Econometrica 5 (3): 280–294

So, really? The academics have known this for seventy-six years and the traders have never cottoned on?

Cowles and Jones used the word "inertia" and said:
The term "inertia," as employed by Mr. Snyder, may be said to be macroscopic, whereas its use in the present study is microscopic. He was concerned with the trend over a period of 100 years or more and concluded that all deviations from that trend were nothing more than inconsequential "jiggles." The present analysis, on the other hand, is concerned primarily with evidence as to inertia in movements of a few hours, days, weeks, months, or years, which Mr. Snyder, with his longer-range viewpoint, would designate as mere "jiggles."
In their analysis,
It was found that, for every series with intervals between observa- tions of from 20 minutes up to and including 3 years, the sequences outnumbered the reversals. For example, in the case of the monthly series from 1835 to 1935, a total of 1200 observations, there were 748 sequences and 450 reversals. That is, the probability appeared to be .625 that, if the market had risen in any given month, it would rise in the succeeding month, or, if it had fallen, that it would continue to decline for another month. The standard deviation for such a long series constructed by random penny tossing would be 17.3; therefore the deviation of 149 from the expected value 599 is in excess of eight times the standard deviation. The probability of obtaining such a result in a penny-tossing series is infinitesimal.
Here's their chart:

Image

But the paper continues--one almost can imagine they are salivating
This evidence of structure in stock prices suggests alluring possibilities in the way of forecasting.
They proceed to spend 286-294 exploring, in detail, various proposed systems for exploiting this evidence of structure, concluding at the end--with what I think is real regret--that none of them worked. With admirable integrity, they point out that
...the results should be interpreted with caution in view of the fact that various units of time, other than 1 month, were considered and rejected. In all, 26 such units, ranging from 20 minutes up to 10 years, were examined. The series represented by units of one month, therefore, was selected by hindsight as the most favorable one in 26 trials.
How often do financial authors showing backtested results come clean on all the tests that were conducted and not presented? They conclude, with what sounds like genuine regret, that
This type of forecasting could not be employed by speculators with any assurance of consistent or large profits.
Last edited by nisiprius on Sat Apr 20, 2013 6:44 am, edited 3 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Dale_G
Posts: 3111
Joined: Tue Feb 20, 2007 5:43 pm
Location: Central Florida - on the grown up side of 81

Re: Momentum based investing in real world

Post by Dale_G » Fri Apr 19, 2013 6:07 pm

asset_chaos wrote:I still have no understanding of what the risk is that should generate a momentum return, while cheerfully stipulating that the data is unambiguous that momentum exists.
The risk is that there will be no "greater fools" around to buy when the trend reverses. Internet stocks had great momentum from about 1997 to 2000. Real estate had great momentum from 2000 to 2007. Momentum is probably great if you can figure out when to leave the party.

Dale
Volatility is my friend

Jebediah
Posts: 515
Joined: Tue Aug 28, 2012 9:19 pm
Location: Denver, CO

Re: Momentum based investing in real world

Post by Jebediah » Fri Apr 19, 2013 6:14 pm

Larry

Your article says:

For value-tilted portfolios, exposure to momentum can provide an important diversification benefit, because momentum is negatively correlated with value.

I'm looking at this chart and MOM seems to track value quite nicely, not inversely. What am I missing?

Image

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Momentum based investing in real world

Post by larryswedroe » Fri Apr 19, 2013 7:22 pm

asset chaos
The tax efficiency is not as bad as one thinks as you have lots of LT gains and lots of ST losses. Toby Moskowitz has paper on this I believe

Jebediah
Yes the correlation is strongly negative, since 1927 monthly its about -0.4 and annual about -0.25
Larry

asset_chaos
Posts: 1388
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: Momentum based investing in real world

Post by asset_chaos » Fri Apr 19, 2013 9:51 pm

nisiprius wrote:So, really? The academics have known this for seventy-six years and the traders have never cottoned on?
Indeed. I had thought (guessed, really) that the anomaly existed because it was too costly to exploit. But Larry's data on a live strategy suggests that too costly is not the case. Being a well-known anomaly and still existing implies that there's a real economic risk that's generating a risk premium to pay people to bear this risk. Though I'm baffled as to what that risk would be. I don't think I buy Dale_G's theory because momentum measurably exists in the market average time series, e.g. the S&P500. So momentum isn't just in popular sectors of the market, it's in every sector of the market and in the total stock markets themselves.

At the end of the day, though, momentum is not something I'll invest in precisely because I don't understand what I would be investing in, what the return source is. More data and much better theory may one day change my mind, but I'm still a long way from that day. More for the bold, better informed investor, I guess.
Regards, | | Guy

Jebediah
Posts: 515
Joined: Tue Aug 28, 2012 9:19 pm
Location: Denver, CO

Re: Momentum based investing in real world

Post by Jebediah » Fri Apr 19, 2013 10:55 pm

Larry: how do you explain the chart? So you're saying that the value load for this fund is negative and yet it tracks Vanguard Value almost perfectly? Maybe Vanguard's value load and AQR's momentum load are both just really small and they're actually overwhelmed by beta?

umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: MOMENTUM based investing in real world

Post by umfundi » Sat Apr 20, 2013 4:19 am

larryswedroe wrote:Louis
Any well run fund that gives up trying to replicate a benchmark to get more effective trading is going to have large tracking error. That's something you have to learn to live with if you use funds like DFA or Bridgeway. They and their investors accept the RANDOM tracking error to improve expected returns. What you have to look for is are the tracking errors truly random. For that you have to do what is called an attribution analysis, which the retail investor is never going to get, or not likely. It's one of the reasons why funds like DFA don't work with public. Trying to explain that issue and having investors understand and accept it is not easy and very time consuming

I hope that is helpful
Larry
Larry,

And this fits with the maxim, do not invest in anything you do not understand?

If you do not mind, I think your post is incredibly condescending.

Keith
Déjà Vu is not a prediction

User avatar
nisiprius
Advisory Board
Posts: 36864
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: MOMENTUM based investing in real world

Post by nisiprius » Sat Apr 20, 2013 6:08 am

umfundi wrote:
larryswedroe wrote:...For that you have to do what is called an attribution analysis, which the retail investor is never going to get, or not likely. It's one of the reasons why funds like DFA don't work with public. Trying to explain that issue and having investors understand and accept it is not easy and very time consuming...
And this fits with the maxim, do not invest in anything you do not understand?
Larry, have you explained "attribution analysis" in any of your books? Is Wikipedia's article on "performance attribution" a reasonably accurate description?

Is attribution analysis a tool that can be used to improve performance, or does it only measure performance after the fact?

If the description is correct--
Attribution analysis attempts to distinguish which of the two factors of portfolio performance, superior stock selection or superior market timing, is the source of the portfolio’s overall performance. Specifically, this method compares the total return of the manager’s actual investment holdings with the return for a predetermined benchmark portfolio and decomposes the difference into a selection effect and an allocation effect.
--then it sounds as if we are talking about active management.

As I continue to Google, I found this! We have at least one skilled practitioner of attribution analysis in our forum: How did you learn about investing and the Bogle way?
Rick Ferri wrote:I converted back in 1996. I was a portfolio management and investment analyst in the brokerage industry at the time. Part of my job was to do in-depth attribution analysis of active movement managers to determine what makes them tick and assess the probability that they may be successful in the future. After several years of hands-on education about how things are actually done on the active side, I converted to index funds in 1996.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Momentum based investing in real world

Post by larryswedroe » Sat Apr 20, 2013 8:09 am

Jebediah
as we know, correlations drift and the figures are the long term results.

Keith
Sorry you found it condescending. Certainly not meant to be. And yes MY RULE is that if you don't fully understand something you should not invest because when the risks show up you will bail out. Like with these funds. They accept tracking error to gain long term advantages. For example, DFA's funds screen out certain stocks (negative momentum, low priced, utilities, reits and so on) and do block trading that make them look very different than an index. That error can be as much as 8% a year and can average a few percent a year. It can be result of things that are difficult to explain to investors. That is actually one of the main reasons DFA doesn't accept retail funds. Now you can think that is condescending, but it's reality. Perhaps after you read my explanation of DFA LV fund performance in 2002 you will change your mind.

Nisiprius
No did not discuss it in my books and it's only for after the fact analysis--and its very different than what Rick was talking about because it has no predictive value. What it does tell you is did the fund execute the strategy or did it drift and what led to the returns it got that were different than its benchmarks

So here are some examples.
Say DFA sv fund underperforms a benchmark like a R2k v. Attribution analysis would attempt to find the explanations by looking at things like
a) how did DFAs holdings differ from the index's by industry, sector. For example we know DFA doesn't hold utilities or REITs, so that could explain it.
b) it could be that a DFA TM fund was holding stocks that had ST gains and was waiting to sell for LT gains and those stocks randomly either outperformed or underperformed--and their over weighting led to the TE
c) DFA has buy range but also hold range for stocks an index would sell--the stocks in the hold range could under or outperform
d) DFA reconstitutes buy and hold ranges all the time, vs index once a year. So as stocks drift into and out of the range, DFA maintains more constant exposure to the factors. So if a factor underperforms during the year, or during even part of year when cash flows come in, then it could underpeform the index. Or reverse.

Here is great example that led to two DFA funds in I think 2002 have large gap in returns, even though basically same fund. One was TM and one was not. Think it was the LV funds.
DFA had recently created the TM LV fund or Marketwide value. So it was generating lots of new cash. This was before DFA implemented momentum screens (and was the final straw that got Fama and French to throw in the towel and use MOM, at least IMO). At the same time their existing LV fund was not only not getting cash but losing it to the new TM fund.

Now what happened next was that there were lots of telecom stocks that were once growth stocks that were dropping into the value buy range. Since DFA focuses on trading costs it won't sell existing stocks to buy new stocks that have become eligible to buy. It waits for cash flows or has sell opportunities as stocks migrate out. So the regular fund did not buy these telecom stocks that kept going down and down. But the TM fund did of course as it had the cash. So the regular fund far outperformed the TM fund. Only an attribution analysis would reveal the explanation.

I think most Bogleheads would be shocked at how large the random tracking errors can be
2002
The FF large value ex utilities index -30.3%, the FF large value with utilities index did much better, down just 20.4%. And the FF large value research index, which is closer to how DFA runs their fund lost 25%
Now the impact of the telecom stocks was so great that it explained most of the difference you will now see
DFA regular large value fund lost just 14.9%, a great year in relative terms, far outperforming its benchmark. Of course no one ever complains or even asks about positive tracking error.
However the TM fund which was buying those falling telecom stocks lost 27.2%, underperforming the non TM version of the same fund by 12.3%. That's random tracking error

There were other issues like industry weightings and so on that also impacted returns, as they do every year. Now does anyone really think you want to try and explain this type stuff to the average retail investor.
We spent several hours going over this with DFA to make sure we understood it. Now they had to do that with of course hundreds of other firms. Imagine doing it with the public on 800 call in lines?

Here is another recent example. You have a really small value fund, smaller than its benchmarks. So it underperforms. You do attribution analysis and you find it that randomly midcap values outperformed small cap value, so that explains the difference

I hope that is helpful

Larry

User avatar
nisiprius
Advisory Board
Posts: 36864
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Momentum based investing in real world

Post by nisiprius » Sat Apr 20, 2013 8:45 am

Larry, you are talking about this?

Image

And attribution analysis shows that the reason was that the older fund, DFA U.S. Large-Cap, DFLVX value, established 1993, was in a stable state of being chronically somewhat strapped for cash. So, it waited to track the index for a few key months that randomly happened to matter? While the newer tax-managed fund, DFA Tax-Managed US Marketwide Value, DTMMX, created in 1999 was growing flush with cash and tracked it?

And it was just the luck of the draw, and could have gone the other way--if the index had dipped, the newer fund would have dipped with it while the older one, by delaying, would have made a lucky, accidentally successful market-timing move?

And when we add the Vanguard [Large] Value index fund, VIVIX, green line...

Image

...it shows why DFLVX should be happy DFA invests the way it does and shouldn't complain that another similar DFA fund had accidental positive tracking error?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Momentum based investing in real world

Post by larryswedroe » Sat Apr 20, 2013 9:22 am

Nisiprius
Of course the outcomes were pure luck. That is exactly the point, they were random outcomes. But do you think individual investors would understand that, or react to the returns differences?
Also it is important to understand exactly what caused the differences because you can then and only then know if the fund is doing what it is supposed to do, following its mandate and rules, and not starting to drift from its passive approach. Without doing the analysis all the time you cannot know that. That is why it's important to do attribution analysis. We go over these issues with fund companies whenever there is significant TE to make sure we understand what is happening, that TE is either random, or result of smart strategies like patient trading.

As to the older fund it was not getting much new cash because investors were moving asset from it to the TM fund (now that K gains were not as much of an issue with the bear market) and new cash was going to the TM fund. But the TM fund was gaining cash.

The key was to understand that there was nothing wrong with the methodology for the DFA TM fund, just bad random luck and there was nothing great about the non TM fund, just random good luck. This shows how much of a role serendipity can play when you don't try to replicate an index to avoid the negatives of pure indexing.

Larry

User avatar
nisiprius
Advisory Board
Posts: 36864
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Momentum based investing in real world

Post by nisiprius » Sat Apr 20, 2013 9:29 am

Wasn't arguing, just making sure I understood you, by creating an illustrated version of your narrative. The question marks were not intended to communicate skepticism, but uncertainty as to whether I was accurately reproducing your points. If I've gotten it correctly, this particular story seems like a nice illustration of DFA's approach.
But do you think individual investors would understand that, or react to the returns differences?
"The things that made your blue line fall below the orange line are the same things that made your blue line fall above the green line?"
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

Carpe
Posts: 211
Joined: Tue Jun 16, 2009 3:38 pm

Re: MOMENTUM based investing in real world

Post by Carpe » Sat Apr 20, 2013 9:39 am

larryswedroe wrote:Louis
Any well run fund that gives up trying to replicate a benchmark to get more effective trading is going to have large tracking error. That's something you have to learn to live with if you use funds like DFA or Bridgeway. They and their investors accept the RANDOM tracking error to improve expected returns. What you have to look for is are the tracking errors truly random. For that you have to do what is called an attribution analysis, which the retail investor is never going to get, or not likely. It's one of the reasons why funds like DFA don't work with public. Trying to explain that issue and having investors understand and accept it is not easy and very time consuming

I hope that is helpful
Larry
No brownie points for DFA and their clients for accepting RANDOM tracking error. It is what is expected of a good index tracker. It is what is expected with Vanguard.
Carpe: pick, pluck, pluck off, gather

stlutz
Posts: 4784
Joined: Fri Jan 02, 2009 1:08 am

Re: Momentum based investing in real world

Post by stlutz » Sat Apr 20, 2013 9:44 am

No brownie points for DFA and their clients for accepting RANDOM tracking error. It is what is expected of a good index tracker. It is what is expected with Vanguard.
In fairness, this kind of error has occurred in the past with VG when they switched index providers from S&P to MSCI. The exact timing of the switch ended up hurting the returns of their value index funds and helping the returns of their growth funds. Will be interesting to see what the impact of the CRSP switch will be...

afan
Posts: 3879
Joined: Sun Jul 25, 2010 4:01 pm

Re: Momentum based investing in real world

Post by afan » Sat Apr 20, 2013 7:59 pm

A better indicator of the fidelity of tracking the target index is R^2, rather than return over 3 years. Plenty of funds may have given returns similar to the AQR offerings, while not at all using momentum strategies. Even better would be a multiple correlation showing the factor loadings of the momentum funds. If this is high, and there is a high correlation with the momentum factor (sort of the same thing), then the strategy is implementable.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Momentum based investing in real world

Post by larryswedroe » Sun Apr 21, 2013 8:04 am

Carpe
With index funds there is pretty much NO tracking error by definition. They do everything to avoid it as their role is to TRACK or REPLICATE the index.
DFA and other passive funds accept TE as the price you pay for avoiding the negatives of indexing, and TE can be very large, much larger than I would bet the Bogleheads would have thought, as my example shows.
Now most investors when they see POSITIVE TE they just ignore it, never even questioning it (not good idea because if there can be good TE there can be bad TE). But when they see negative TE they think it's fund's fault and bail, assets leave and trading costs go up (it's usually at time when liquidity cost is high) and tax efficiency goes down as well.
Best wishes
Larry

User avatar
nedsaid
Posts: 10499
Joined: Fri Nov 23, 2012 12:33 pm

Re: Momentum based investing in real world

Post by nedsaid » Sun Apr 21, 2013 9:45 am

There is a well known fund company that has used both earnings acceleration and momentum strategies. I have owned some of these over the years, pretty much to put a "tiger in my tank." This fund company has emphasized earnings acceleration in its older more traditional funds. The company offered newer funds that had a greater emphasis on momentum over earnings acceleration, the newer funds did not perform as well. The older funds are high turnover and the newer funds have even higher turnover. I passed on the newer style funds.

All the technical analysis and all the computers grinding away 24/7 did not enhance the performance of the newer funds (more momentum oriented) over the older funds (more earnings acceleration oriented). The newer funds have less input from the managers than the older funds have.

Earnings acceleration works brilliantly sometimes, you have to accept that sometimes it doesn't do too well. When it does well, WOW. This company built its success on the old adage that money follows earnings. The momentum based strategies based on price and volume don't seem to have added value there.

It is a classic case of a strategy looking good on paper and failing to meet expectations in real life.
A fool and his money are good for business.

Post Reply