WSJ: Most market anomalies academics have identified don’t exist

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

WSJ: Most market anomalies academics have identified don’t exist

Post by backpacker » Thu May 11, 2017 3:30 pm

From the Wall Street Journal, a summary of new study that attempted to replicate 447 academic factors using consistent statistical methodology. A link to the original article "An Algorithm, an ETF and an Academic Study Walk Into a Bar" can be found here.

As it turns out, 380 factors failed replication and many of the remaining proved to be much smaller than originally estimated.
A new study making waves in quantitative finance tested 447 anomalies identified by academics and found more than eight out of 10 vanish when rigorous tests are applied. Among those failing to reach statistical significance: one anomaly recently set out by the godfathers of quantitative finance, Nobel-winning economist Eugene Fama and his colleague Kenneth French.

The study, “Replicating Anomalies,” published this week by Kewei Hou and Lu Zhang at Ohio State University and Chen Xue at the University of Cincinnati, is the biggest test of examples of inefficient markets carried out so far. The trio applied consistent analysis to the supposed anomalies, used the same database of stocks and set higher standards for statistical significance. Simply reducing the influence of the plethora of rarely traded penny stocks—which make up just 3% of market value but 60% of all listings—by using market capitalization weightings made more than half of past findings no longer significant.

Messrs. Hou, Xue and Zhang warn that academics have been fiddling the statistics to come up with interesting findings, known to statisticians as data mining or p-hacking. “The anomalies literature is infested with widespread p-hacking,” they write.

It isn’t all bad news for investors and those trying to make a living flogging what have become known as “factors.” The research confirmed that the most popular factors have indeed outperformed the market over long periods even when faced with rigorous tests, but found much smaller returns than previous studies estimated.

User avatar
G-Force
Posts: 276
Joined: Sun Sep 20, 2009 3:03 pm
Location: Florida
Contact:

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by G-Force » Thu May 11, 2017 3:34 pm

Non paywall version: https://archive.is/QvA4Z

Whakamole
Posts: 466
Joined: Wed Jan 13, 2016 9:59 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Whakamole » Thu May 11, 2017 3:37 pm

I like the quote from Asness saying he's worried people will start discarding factors. Because that's the most important thing, we don't want people dismissing smart beta.

livesoft
Posts: 56519
Joined: Thu Mar 01, 2007 8:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by livesoft » Thu May 11, 2017 3:45 pm

Lucky for us, they confirmed the factors that people at bogleheads.org use as being quite robust.
This signature message sponsored by sscritic: Learn to fish.

grok87
Posts: 7567
Joined: Tue Feb 27, 2007 9:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grok87 » Thu May 11, 2017 4:31 pm

Whakamole wrote:I like the quote from Asness saying he's worried people will start discarding factors. Because that's the most important thing, we don't want people dismissing smart beta.
I thought that was funny too

Personally i think momentum is a crock. Picking up nickels in front of a steamroller. The momemtum factor lost 83% in 2009. You have to go back 40 years before the average annual return is materially positive and that's ignoring trading costs.
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

User avatar
triceratop
Moderator
Posts: 3609
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by triceratop » Thu May 11, 2017 4:41 pm

livesoft wrote:Lucky for us, they confirmed the factors that people at bogleheads.org use as being quite robust.
Are you looking at the same small-cap plot I am looking at?
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

lack_ey
Posts: 5698
Joined: Wed Nov 19, 2014 11:55 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by lack_ey » Thu May 11, 2017 5:05 pm

grok87 wrote:Personally i think momentum is a crock. Picking up nickels in front of a steamroller. The momemtum factor lost 83% in 2009. You have to go back 40 years before the average annual return is materially positive and that's ignoring trading costs.
What about on the long side only, or scaling for volatility, or looking outside stocks?

Dominic
Posts: 120
Joined: Sat Jul 02, 2016 11:36 am

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Dominic » Thu May 11, 2017 5:49 pm

grok87 wrote:
Whakamole wrote:I like the quote from Asness saying he's worried people will start discarding factors. Because that's the most important thing, we don't want people dismissing smart beta.
I thought that was funny too

Personally i think momentum is a crock. Picking up nickels in front of a steamroller. The momemtum factor lost 83% in 2009. You have to go back 40 years before the average annual return is materially positive and that's ignoring trading costs.
I have a feeling momentum is going to perform poorly. I don't see why the market should reward investors who buy assets that have already been successful. These stocks shouldn't be more risky than others. Maybe I'm wrong, but I think it's mostly a behavioral issue that the market will correct.

I'm currently not a factor investor because my portfolio is too small to make it worth my while to invest in anything other than an all-in-one fund. As I save more, I plan to tilt to small cap value. That's a factor combination with crystal clear risk story, and the three-factor model does a very good job of explaining equity returns in my opinion.

The "factors" in this study, however, are data-mined and I'm not surprised they aren't replicable.

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Thu May 11, 2017 5:57 pm

Larry's criteria for a factor are: persistent, pervasive, robust, intuitive, investable. I think intuitive is very important here. When articles like this or long periods of underperformance occur, I think it's really important to have a strong belief in a rational risk or behavioral story to make one stick with his plan. Pervasive, persistent, and robust all look backwards. Intuitive, I believe, really helps to look forwards.

Dave

grok87
Posts: 7567
Joined: Tue Feb 27, 2007 9:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grok87 » Thu May 11, 2017 5:59 pm

lack_ey wrote:
grok87 wrote:Personally i think momentum is a crock. Picking up nickels in front of a steamroller. The momemtum factor lost 83% in 2009. You have to go back 40 years before the average annual return is materially positive and that's ignoring trading costs.
What about on the long side only, or scaling for volatility, or looking outside stocks?
long only sounds potentially less dangerous, but i don't know.

IMHO "scaling for volatility" is not going to work. It reminds me of the portfolio insurance idea that didn't protect again losses in the 1987 crash and arguably was one of the factors that caused the crash.
http://www.nytimes.com/2012/10/19/busin ... treet.html
When the monsters come, "it ["portfolio insurance", "volatility scaling"] won''t make any difference."
https://www.youtube.com/watch?v=q9yAfQ3CTnc


i'm not sure if looking outside stocks would help. Are you meaning that there might be some diversification benefit from playing momentum in different asset classes (the Asness argument) or that the tails might be less fat?
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Thu May 11, 2017 6:36 pm

Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave

grok87
Posts: 7567
Joined: Tue Feb 27, 2007 9:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grok87 » Thu May 11, 2017 6:42 pm

Random Walker wrote:Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave
thanks Dave.
Well the fund hasn't really been through a crisis yet. I guess we'll see then if that diversification helps or not.
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

livesoft
Posts: 56519
Joined: Thu Mar 01, 2007 8:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by livesoft » Thu May 11, 2017 6:46 pm

Bloomberg's Eric Weiner reports as well: https://www.bloomberg.com/news/articles ... -imaginary
This signature message sponsored by sscritic: Learn to fish.

slowmoney
Posts: 164
Joined: Tue Dec 28, 2010 4:41 pm
Location: Louisiana Purchase

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by slowmoney » Thu May 11, 2017 7:04 pm

Random Walker wrote:
When articles like this or long periods of underperformance occur, I think it's really important to have a strong belief in a rational risk or behavioral story to make one stick with his plan. Pervasive, persistent, and robust all look backwards. Intuitive, I believe, really helps to look forwards.
This is a great point.

As an appraiser, you quickly realize that the past has very little to do with the market going forward. What prices were last year, what was hot or cold, etc., has no bearing on what people are currently willing to pay. It only matters what homes, trends, plant closures, etc are impacting the market right now. That is the price. Your opinion or what happened last year, does not matter. IMO too many investors look to the past because it's easy. Looking forward requires thinking, guessing, courage and the knowledge that your probably wrong.

I have enough trouble "staying the course" without years and years of factor underperformance.

Real question: How can a factor be persistent, robust and pervasive and yet suffer years, maybe decades of underperformance?
Information is more valuable sold than used. - Fischer Black (1938-1995)

User avatar
nedsaid
Posts: 8707
Joined: Fri Nov 23, 2012 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by nedsaid » Thu May 11, 2017 7:19 pm

Simply reducing the influence of the plethora of rarely traded penny stocks—which make up just 3% of market value but 60% of all listings—by using market capitalization weightings made more than half of past findings no longer significant.
Interesting, if you just take what I call the "anti-factors" out, or those stocks with negative characteristics, you will get good performance. The indexes screen out the worst of the "anti-factor" stocks, hence Vanguard Small Value Index and Vanguard Small Growth Index perform in similar factors. These are the "lottery stocks" that Larry Swedroe talks about. I have wondered if those awful lottery stocks are sort of like the bogeyman who is going to get you.

The article also said that even with the more rigorous testing that Value, Momentum, Quality still outperform and that stocks with high investment underperform. It seems that the premiums are less than previously thought and it matters how you screen for factors.

I remember all the discussions about Vanguard Small Cap Value Index and whether it sufficiently loaded on the Value and Size factors. These threads popped up after I purchased this index in ETF form. If you look at the Vanguard product through the Morningstar X-Ray, you see a lot of Mid-Cap stocks and a lot of stocks that aren't Value in the index. Also, I remember that Morningstar criticized a Micro-Cap Index ETF that I purchased. Front running was and is a problem with the Micro-Cap Index ETF. (Now they tell me!) So the construction of the investment products you purchase has a lot to do with whether you capture those factors or not.

So I don't know. It reminds me of Nedsaid's Rule of Investing number 34. Here it is below:
Once you make a big investment decision, I can guarantee that you will see something in the financial media that will tell you that you have made a big mistake. Also keep in mind the media is relentlessly pessimistic and this will lead you to believe that things are worse than they really are.
A fool and his money are good for business.

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Thu May 11, 2017 7:22 pm

Showmoney,
On Monday May 8 I posted an article written by Larry Swedroe on Factor diversification. There are two tables in that article that make a strong start towards answering your question.

http://www.etf.com/sections/index-inves ... sification

Dave

slowmoney
Posts: 164
Joined: Tue Dec 28, 2010 4:41 pm
Location: Louisiana Purchase

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by slowmoney » Thu May 11, 2017 7:41 pm

Thanks for the reply, Random Walker. :D

However, I guess I remain unconvinced. Being unconvinced is a bad place from which to "stay the course".
Information is more valuable sold than used. - Fischer Black (1938-1995)

lack_ey
Posts: 5698
Joined: Wed Nov 19, 2014 11:55 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by lack_ey » Thu May 11, 2017 7:55 pm

grok87 wrote:
lack_ey wrote:
grok87 wrote:Personally i think momentum is a crock. Picking up nickels in front of a steamroller. The momemtum factor lost 83% in 2009. You have to go back 40 years before the average annual return is materially positive and that's ignoring trading costs.
What about on the long side only, or scaling for volatility, or looking outside stocks?
long only sounds potentially less dangerous, but i don't know.
grok87 wrote:IMHO "scaling for volatility" is not going to work. It reminds me of the portfolio insurance idea that didn't protect again losses in the 1987 crash and arguably was one of the factors that caused the crash.
http://www.nytimes.com/2012/10/19/busin ... treet.html
When the monsters come, "it ["portfolio insurance", "volatility scaling"] won''t make any difference."
https://www.youtube.com/watch?v=q9yAfQ3CTnc
Vol scaling is very different functionally and theoretically. Anyway, the point here is not that it's always going to help, just that it might sometimes, and with good reason. Usually when markets are more volatile, they're more likely to do something extreme either on the upside or downside. Same for factors. Sometimes you get 1987 which came out of a medium volatility environment but frequently there are more rumblings beforehand, higher realized vol, like 2008-2009. Anyway, it was mentioned in part because it would have prevented losses of that magnitude for a long/short factor formulation.

I mean, if losing stocks are plunging A LOT, you should not be surprised if a momentum strategy shorting them gets wrecked if these stocks rebound. Vol scaling means that if things are going nuts, maybe you're not 100% short stocks that are jumping around like crazy. It's a tail-reducing measure, for both tails.
grok87 wrote:i'm not sure if looking outside stocks would help. Are you meaning that there might be some diversification benefit from playing momentum in different asset classes (the Asness argument) or that the tails might be less fat?
Both, though I meant more of the former.

The key is that you don't want to be 100% short stocks based on one factor/strategy, no matter what it is, especially when stocks are doing crazy things. Especially not momentum, though, I'll give you that.

It's just that quoting one vanilla momentum (unscaled, etc.) crash in equities is a bit alarmist out of context and doesn't encompass the whole of what momentum behavior is like generally. It's even worse than just bringing up 1929-1932 every time somebody mentions stock investing. Here nobody really invests using the momentum factor portfolio, unlike investing in the stock market as a whole (which people do).

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Thu May 11, 2017 8:42 pm

Slomoney,
Being unconvinced can create a lot of conviction to stay the course with a low cost TSM approach; hard to go wrong with that! I first got interested in Boglehead efficient market hypothesis investing because I saw so many parallels between perfect competition from economics in college and equilibrium from chemistry. With those in my head TSM was strongly engrained. Not sure how I evolved to a factor junkie :-)

Dave

User avatar
HomerJ
Posts: 10369
Joined: Fri Jun 06, 2008 12:50 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by HomerJ » Thu May 11, 2017 9:02 pm

grok87 wrote:
Random Walker wrote:Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave
thanks Dave.
Well the fund hasn't really been through a crisis yet. I guess we'll see then if that diversification helps or not.
The last two times AQR went through a crisis (2000 and 2008), some of their other funds almost went bankrupt. Cliff admitted he was 3 months from closing shop in 2000 if the market hadn't finally crashed (he was short a lot of tech stocks). In 2008, his defense was that he had never seen a market like that. It was outside any models they had.

Those funds were also hedge funds. Invested in multiple factors. With models based on historical data.

I hope the next crisis is something we've already seen, because then it will be accounted for in his models. If it's something new, and/or it lasts 3 months too long, you may lose a ton of money investing with him. It's certainly not riskless, even though he's selling it that way because "diversification"

garlandwhizzer
Posts: 1621
Joined: Fri Aug 06, 2010 3:42 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by garlandwhizzer » Thu May 11, 2017 9:12 pm

I'm with grok87 on this one. I believe the MOM premium is very difficult to harvest after costs and most who try will fail. Long/short MOM as grok87 has pointed out tends to suffer massive losses, the kind that takes long years to recover from, in the recoveries following major market collapses. Huge losses happen very quickly and severely in such situations even as the rest of the market is going up rapidly. These losses are heavily concentrated in the short side of the portfolio. If you wish to play the MOM game, I suggest long only large cap exposure. Trading costs and frictions in LC don't devour returns like they do in smaller cap spaces. Long only LC MOM, like MTUM, offers diversification to SCV as well as the possibility (but not the certainty) of some outperformance relative to beta index portfolios. I do not do that myself but I believe it to be a reasonable choice for those who desire multi-factor exposure to Sm,Val, and MOM.

Garland Whizzer

Whakamole
Posts: 466
Joined: Wed Jan 13, 2016 9:59 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Whakamole » Fri May 12, 2017 1:17 pm

HomerJ wrote:
grok87 wrote:
Random Walker wrote:Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave
thanks Dave.
Well the fund hasn't really been through a crisis yet. I guess we'll see then if that diversification helps or not.
The last two times AQR went through a crisis (2000 and 2008), some of their other funds almost went bankrupt. Cliff admitted he was 3 months from closing shop in 2000 if the market hadn't finally crashed (he was short a lot of tech stocks). In 2008, his defense was that he had never seen a market like that. It was outside any models they had.

Those funds were also hedge funds. Invested in multiple factors. With models based on historical data.

I hope the next crisis is something we've already seen, because then it will be accounted for in his models. If it's something new, and/or it lasts 3 months too long, you may lose a ton of money investing with him. It's certainly not riskless, even though he's selling it that way because "diversification"
Or will be something else his models haven't predicted because it hasn't happened before. Wasn't that the mistake that brought down LTCM?

grok87
Posts: 7567
Joined: Tue Feb 27, 2007 9:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grok87 » Fri May 12, 2017 6:55 pm

Whakamole wrote:
HomerJ wrote:
grok87 wrote:
Random Walker wrote:Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave
thanks Dave.
Well the fund hasn't really been through a crisis yet. I guess we'll see then if that diversification helps or not.
The last two times AQR went through a crisis (2000 and 2008), some of their other funds almost went bankrupt. Cliff admitted he was 3 months from closing shop in 2000 if the market hadn't finally crashed (he was short a lot of tech stocks). In 2008, his defense was that he had never seen a market like that. It was outside any models they had.

Those funds were also hedge funds. Invested in multiple factors. With models based on historical data.

I hope the next crisis is something we've already seen, because then it will be accounted for in his models. If it's something new, and/or it lasts 3 months too long, you may lose a ton of money investing with him. It's certainly not riskless, even though he's selling it that way because "diversification"
Or will be something else his models haven't predicted because it hasn't happened before. Wasn't that the mistake that brought down LTCM?
it's even worse than that.

whatever assumptions he and others build into their models and upon which they have constructed their defenses will be the very thing to bring them down.

for example the volatility targeting thing. the assumption is that rising volatility is a signal of coming risk. And their defense is then to start selling, scaling down their positions to manage risk. the trouble is everyone else has the same plan so everyone is going to sell at the same time. the rout will then accelerate in a self-reinforcing downward spiral.

take the run up to the global financial crisis. A massively leveraged financial edifice was built on the assumption that there would never be a nationwide downturn in housing prices. And then guess what happened? (Answer a massive nationwide downturn in housing prices).

Investing is not physics. the bad stuff that happens are not random external shocks. the bad stuff that happens is almost always of our own design. We have met the enemy and he is us.
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

User avatar
nedsaid
Posts: 8707
Joined: Fri Nov 23, 2012 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by nedsaid » Fri May 12, 2017 7:21 pm

grok87 wrote:
Whakamole wrote:
HomerJ wrote:
grok87 wrote:
Random Walker wrote:Grok,
Yes there is Momentum everywhere. The AQR funds invest in equities, bonds, commodities, currencies. Diversifying across asset classes within the Style improves its efficiency lots.

Dave
thanks Dave.
Well the fund hasn't really been through a crisis yet. I guess we'll see then if that diversification helps or not.
The last two times AQR went through a crisis (2000 and 2008), some of their other funds almost went bankrupt. Cliff admitted he was 3 months from closing shop in 2000 if the market hadn't finally crashed (he was short a lot of tech stocks). In 2008, his defense was that he had never seen a market like that. It was outside any models they had.

Those funds were also hedge funds. Invested in multiple factors. With models based on historical data.

I hope the next crisis is something we've already seen, because then it will be accounted for in his models. If it's something new, and/or it lasts 3 months too long, you may lose a ton of money investing with him. It's certainly not riskless, even though he's selling it that way because "diversification"
Or will be something else his models haven't predicted because it hasn't happened before. Wasn't that the mistake that brought down LTCM?
it's even worse than that.

whatever assumptions he and others build into their models and upon which they have constructed their defenses will be the very thing to bring them down.

for example the volatility targeting thing. the assumption is that rising volatility is a signal of coming risk. And their defense is then to start selling, scaling down their positions to manage risk. the trouble is everyone else has the same plan so everyone is going to sell at the same time. the rout will then accelerate in a self-reinforcing downward spiral.

take the run up to the global financial crisis. A massively leveraged financial edifice was built on the assumption that there would never be a nationwide downturn in housing prices. And then guess what happened? (Answer a massive nationwide downturn in housing prices).

Investing is not physics. the bad stuff that happens are not random external shocks. the bad stuff that happens is almost always of our own design. We have met the enemy and he is us.
Grok87, I like what you said that investing is not physics. It reminds me of when Larry Swedroe did a typo and said one of the top people at Buckingham had a PhD in psychics. There were a lot of chuckles over that, and truth be told, Larry's typo contained more truth than we will ever know.

The reality is that the markets have a way of doing the very thing that you never expected. This is why I don't like the leverage and shorting techniques that the hedge funds use. You don't know what the other big market players are doing. You may think yourself the contrarian when in fact a whole bunch of others might be doing exactly what you are doing. This hedge fund stuff gets to be like the spy vs spy and the measure vs countermeasure stuff in the Cold War. What happens is that the "smart people" themselves get outsmarted and perform what seems to us amateurs to be a very fundamental error. Sometimes, "smart money" really does some crazy and stupid stuff and us individual investors look brilliant by comparison.

The thing is, markets are mostly rational but can experience bouts of irrationality. It is human emotion, the greed and fear thing. We are better off having a good understanding of market history, human nature, and human behavior. Trying to reduce all of this to a physics equation is just nuts. There is no theory of everything for investing. Behavioral finance comes the closest.

Of course, some of the quants on this forum think I am an idiot and don't know what I am talking about. They don't understand the limits of accounting, and though the accountants do a good job, the numbers are not exact and contain things like judgment, estimates, and accruals. How can you arrive at precision when the numbers you start out with are not precise themselves? You just cannot explain this to a quant.

I have posted several times about Microsoft. The accountants give it a book value of about $9 a share. The market gives Microsoft stock, last I looked, a market value of about $70 a share. There is a $61 a share difference between what the accountants and what the market values Microsoft. I have explained why this is so and in another thread, Valuethinker made comments along this line as well. And yet the academics use book value per share as one measurement for determining value. Dude, what happened to the $61? This is part of my frustration in trying to explain this stuff to quants.

I am not arguing that book value is a meaningless measurement, I am saying that it has its limitations. As you look at financial data, you realize there is more squishiness to all of this than what is admitted to. Another example of what I am talking about is trying to construct efficient frontiers for your portfolio. Depending on what time periods you use, the efficient frontiers can move all over the place. I am not saying that efficient frontiers are bunk but that there are some pretty severe limitations. An imperfect tool in the toolbox and that is about it.

There gets to be a point where good enough is good enough and close enough is close enough. None of this is precise and even the experts do more eyeballing and educated guesses than what is admitted to. But then again, I am only right about anything around here about every six months or so.
A fool and his money are good for business.

User avatar
knpstr
Posts: 1798
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by knpstr » Fri May 12, 2017 8:23 pm

Most things academics have identified in MPT doesn't exist. Or rather, don't mean (or work out to) what they portray them to mean.

However, the disconnect is not just a phenomena of the finance field.
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

ryman554
Posts: 785
Joined: Sun Jan 12, 2014 9:44 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by ryman554 » Sat May 13, 2017 12:20 am

triceratop wrote:
livesoft wrote:Lucky for us, they confirmed the factors that people at bogleheads.org use as being quite robust.
Are you looking at the same small-cap plot I am looking at?
As far as I know, the *only* factor that bogleheads like to promote with general agreement is "time money is in market", but it's a stretch to call that a factor vs. good investment behaviour.

Everything else seems debatable from my view in the seats...

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by NiceUnparticularMan » Sat May 13, 2017 5:49 am

ryman554 wrote:
triceratop wrote:
livesoft wrote:Lucky for us, they confirmed the factors that people at bogleheads.org use as being quite robust.
Are you looking at the same small-cap plot I am looking at?
As far as I know, the *only* factor that bogleheads like to promote with general agreement is "time money is in market", but it's a stretch to call that a factor vs. good investment behaviour.

Everything else seems debatable from my view in the seats...
I don't know if there is general agreement behind any particular view, if that means 100% agreement. But I would note that anyone who believes we should pick an allocation between stocks and bonds based on the propositions that more stocks add expected return but also risk and more bonds lower expected return but also lower risk is implicitly endorsing something like the one-factor CAPM model.

User avatar
triceratop
Moderator
Posts: 3609
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by triceratop » Sat May 13, 2017 9:41 am

ryman554 wrote:
triceratop wrote:
livesoft wrote:Lucky for us, they confirmed the factors that people at bogleheads.org use as being quite robust.
Are you looking at the same small-cap plot I am looking at?
As far as I know, the *only* factor that bogleheads like to promote with general agreement is "time money is in market", but it's a stretch to call that a factor vs. good investment behaviour.

Everything else seems debatable from my view in the seats...
That is not what livesoft nor I were talking about.

We were talking about academic factors in explaining portfolio returns, as this entire thread is concerned with them.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

freebeer
Posts: 1902
Joined: Wed May 02, 2007 8:30 am
Location: Seattle area USA

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by freebeer » Sat May 13, 2017 10:18 am

Random Walker wrote:Larry's criteria for a factor are: persistent, pervasive, robust, intuitive, investable. I think intuitive is very important here. When articles like this or long periods of underperformance occur, I think it's really important to have a strong belief in a rational risk or behavioral story to make one stick with his plan. Pervasive, persistent, and robust all look backwards. Intuitive, I believe, really helps to look forwards.

Dave
Intuitive may help in sticking to a plan. But it seems like a red herring criteria for a factor because a plausible "just so story" doesn't make something true. That bloodletting removed bad stuff from sick bodies was intuitive enough that people kept using leeches for centuries! But maybe I'm just biased because I've just read Michael Lewis' newest book, The Undoing Project, which is indirectly all about behavioral economics. After reading that book I don't think I'll ever be able to equate "intuitive" with "true".

User avatar
nedsaid
Posts: 8707
Joined: Fri Nov 23, 2012 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by nedsaid » Sat May 13, 2017 11:44 am

freebeer wrote:
Random Walker wrote:Larry's criteria for a factor are: persistent, pervasive, robust, intuitive, investable. I think intuitive is very important here. When articles like this or long periods of underperformance occur, I think it's really important to have a strong belief in a rational risk or behavioral story to make one stick with his plan. Pervasive, persistent, and robust all look backwards. Intuitive, I believe, really helps to look forwards.

Dave
Intuitive may help in sticking to a plan. But it seems like a red herring criteria for a factor because a plausible "just so story" doesn't make something true. That bloodletting removed bad stuff from sick bodies was intuitive enough that people kept using leeches for centuries! But maybe I'm just biased because I've just read Michael Lewis' newest book, The Undoing Project, which is indirectly all about behavioral economics. After reading that book I don't think I'll ever be able to equate "intuitive" with "true".
Investment experience counts for something. I have owned various investments and watched how they performed under different market conditions. As the late Yogi Berra once said, "You can observe a lot by watching." The academic research was intuitive to me in that it was consistent with what I know about human nature and human behavior and was consistent with what I have myself have learned from eyeballing my investments. I learned new things from the research but much of it confirmed what I already knew or suspected. The rest of it made sense from a behavioral point of view. It isn't like it all came from thin air.

The thing is, if the academic research had told us to all buy mining stocks on the Vancouver stock exchange or dabble in penny stocks, I would have just rolled my eyes. In a similar fashion, I don't believe much of anything I see in certain tabloid headlines.

I know what Random Walker is trying to say.
A fool and his money are good for business.

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Sat May 13, 2017 12:15 pm

I'll have to read Lewis' The Undoing Project! Thanks for mentioning the book :-)

Dave

freebeer
Posts: 1902
Joined: Wed May 02, 2007 8:30 am
Location: Seattle area USA

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by freebeer » Sat May 13, 2017 2:00 pm

nedsaid wrote:...The academic research was intuitive to me in that it was consistent with what I know about human nature and human behavior and was consistent with what I have myself have learned from eyeballing...
Yeah that was exactly the line of "reasoning" of all the economists who attacked Tversky and Kahneman for decades for exposing the flaws of models built on assumptions that people behaved rationally. Loved Michael Lewis book about it...

User avatar
nedsaid
Posts: 8707
Joined: Fri Nov 23, 2012 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by nedsaid » Sat May 13, 2017 3:41 pm

freebeer wrote:
nedsaid wrote:...The academic research was intuitive to me in that it was consistent with what I know about human nature and human behavior and was consistent with what I have myself have learned from eyeballing...
Yeah that was exactly the line of "reasoning" of all the economists who attacked Tversky and Kahneman for decades for exposing the flaws of models built on assumptions that people behaved rationally. Loved Michael Lewis book about it...
That is just insulting.

So you read a Michael Lewis book and are now an expert on everything. The thing is, that whatever case you make, there is always a counter argument. Just as there are critics of factor investing, I suppose there are critics of behavioral finance. This stuff gets debated here all the time. But there is no reason to insult the intelligence of people who hold the other point of view.
A fool and his money are good for business.

BigJohn
Posts: 1394
Joined: Wed Apr 02, 2014 11:27 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by BigJohn » Sat May 13, 2017 6:05 pm

nedsaid wrote:The reality is that the markets have a way of doing the very thing that you never expected.
I'm in the middle of reading Taleb's "The Black Swan" and so far this is the essence of what I've taken away regarding it's application to investing. On one hand it worries me as I agree with comments above that the models themselves can create the next big problem. However, since I was never convinced enough to make the leap to any factor investing, my simple 3 fund portfolio is as good as any other choice to ride out whatever problem gets created.

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Sat May 13, 2017 7:11 pm

Isn't Taleb a fan of sort of a barbell approach: combination of riskless and most risky? A ton of T-bills and a dose of the riskiest asset classes?

Dave

freebeer
Posts: 1902
Joined: Wed May 02, 2007 8:30 am
Location: Seattle area USA

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by freebeer » Sat May 13, 2017 7:13 pm

nedsaid wrote:
freebeer wrote:
nedsaid wrote:...The academic research was intuitive to me in that it was consistent with what I know about human nature and human behavior and was consistent with what I have myself have learned from eyeballing...
Yeah that was exactly the line of "reasoning" of all the economists who attacked Tversky and Kahneman for decades for exposing the flaws of models built on assumptions that people behaved rationally. Loved Michael Lewis book about it...
That is just insulting.

So you read a Michael Lewis book and are now an expert on everything. The thing is, that whatever case you make, there is always a counter argument. Just as there are critics of factor investing, I suppose there are critics of behavioral finance. This stuff gets debated here all the time. But there is no reason to insult the intelligence of people who hold the other point of view.
I never said I was an expert on anything nor did I intend to insult your or anyone else's intelligence. The point of Tversky and Kahneman's (Nobel-prize-winning) research was that there are consistent cognitive biases in situation analysis and human decision making that we all (themselves, their critics, everyone) are subject to, and that boils down to - be distrustful of "human intuition". And supporting an academic research result because it comports with that intuition is a prevalent source of systematic error. That includes you, but also me too, so no offense intended.

BigJohn
Posts: 1394
Joined: Wed Apr 02, 2014 11:27 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by BigJohn » Sat May 13, 2017 7:43 pm

Random Walker wrote:Isn't Taleb a fan of sort of a barbell approach: combination of riskless and most risky? A ton of T-bills and a dose of the riskiest asset classes?
I haven't read any specific investment recommendations (yet?). My comment was just based on the principle that it's difficult to predict big events because we are blinded to possibilities that haven't happened yet. Given that, a conservative AA using a 3 fund portfolio without relying on tilts seems to be appropriate, at least for me.

grok87
Posts: 7567
Joined: Tue Feb 27, 2007 9:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grok87 » Sat May 13, 2017 7:52 pm

BigJohn wrote:
Random Walker wrote:Isn't Taleb a fan of sort of a barbell approach: combination of riskless and most risky? A ton of T-bills and a dose of the riskiest asset classes?
I haven't read any specific investment recommendations (yet?). My comment was just based on the principle that it's difficult to predict big events because we are blinded to possibilities that haven't happened yet. Given that, a conservative AA using a 3 fund portfolio without relying on tilts seems to be appropriate, at least for me.
What about a stagflAtion scenario? Not sure how 3 fund protects there? Swensen argues to divide your bonds between treasuries and tips to help protect against deflation scenarios and stagflation scenariOs.
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Sat May 13, 2017 8:01 pm

BigJon,
I believe Taleb is likewise focused on big events, Black Swans, and that is why he advocates a barbell approach. I could be wrong, but pretty sure he uses lots of T bills and some super risky stuff.

Dave

BigJohn
Posts: 1394
Joined: Wed Apr 02, 2014 11:27 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by BigJohn » Sun May 14, 2017 7:22 am

grok, right now I'm relying on stocks to counter inflation so you're right, stagflation could be an issue. It's why I'm considering splitting my bonds allocation and adding TIPS.

RandomWalker, I'm not reading the book for financial advice but more to understand how our brain handles (or does not handle in this case) very low probability events. Two takeaways for me are that we are prone to see patterns were there are none and often judge the future assuming it will repeat the past. Both could well be in play in how we judge factor investing which just reinforced my prior belief that taking this risk is not for me.

Random Walker
Posts: 2045
Joined: Fri Feb 23, 2007 8:21 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by Random Walker » Sun May 14, 2017 8:54 am

BigJohn,
One thing I see all the time in myself and others is we hear 90% and our minds translate that to 100% probability. Larry Swedroe refers to it as treating the likely as certain and the improbable as impossible.

Dave

livesoft
Posts: 56519
Joined: Thu Mar 01, 2007 8:00 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by livesoft » Sun May 14, 2017 9:32 am

Below is a visual demonstration of the power of factor investing.

Yesterday, it was asked on the Trev H Simplified Ultimate thread
robertmcd wrote:Bringing up an old thread, but what would be the stock/bond allocation for the 4 fund ultimate buy and hold that would match the return of 40%TSM/40%Total international/20% ITT? I know it is past performance and all that but I would just like an idea. I am 23 and have a lot of cash to invest. I have great 5 yr CD's available at a local CU that I would like to use to mimic inter term treasuries. I like the idea of 80/20 stock bond split if I were using total market index funds, but I would like to increase my fixed income allocation with my large amount of cash on hand (possibly going to a mortgage payment in the next few years) to keep my stock exposure to a level I am comfortable with.

The 4 funds would be:
vanguard total stock market
vanguard small cap value
vanguard international value
vanguard ex us small cap

Thanks guys.
And Sammy_M did the work:
Sammy_M wrote:^It's going to be very time dependent. From 5/2009 [sic, really 1/2010] to present using equal parts VTI, VBR, VTRIX, VSS - you could have raised your VFIUX allocation from 20 to 32% and achieved the same return with less volatility, i.e. greater sharpe ratio.

https://www.portfoliovisualizer.com/bac ... tion6_2=32
So one portfolio has 20% Vanguard intermediate-term Treasury fund and the other has 32% of that bond fund. That is quite a significant change in the percentage of bonds going from 20% to 32%. Which portfolio would be more risky? The one with less bonds? Here's the chart produced by clicking on Sammy_M's link to portfoliovisualizer.com:

Image

What factors are at work here?
This signature message sponsored by sscritic: Learn to fish.

User avatar
nedsaid
Posts: 8707
Joined: Fri Nov 23, 2012 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by nedsaid » Sun May 14, 2017 10:14 am

freebeer wrote:


I never said I was an expert on anything nor did I intend to insult your or anyone else's intelligence. The point of Tversky and Kahneman's (Nobel-prize-winning) research was that there are consistent cognitive biases in situation analysis and human decision making that we all (themselves, their critics, everyone) are subject to, and that boils down to - be distrustful of "human intuition". And supporting an academic research result because it comports with that intuition is a prevalent source of systematic error. That includes you, but also me too, so no offense intended.
Thanks for the clarification. I appreciate that.

Cognitive bias is a difficult problem to overcome. Larry Swedroe had a thread or two on this, there are studies showing that algorithms make more accurate decisions than humans, one study had to do with medical diagnosis. As far as the academic research, I have read John Bogle's Tell-Tale Chart speech given at a Morningstar conference, posts here and articles that debunk factors. I also understand that Warren Buffett and Charlie Munger are pretty disdainful of the academic research. So I look at both sides of the issue.

There is also getting to be a big group think problem. It seems that fewer people can perform critical thinking. Many now can't look at both sides of the issue and describe the strong and weak arguments for each side. There is something about seeing that the other side has a point. So a lot of discussion gets to be pretty one sided.

I have been reading that a lot of scientific studies cannot be replicated. It seems that money can buy whatever results you want and this is getting to be a real problem. I am also seeing that conventional wisdom that was derived from earlier studies, particularly related to human health, is now being challenged.

Healthy skepticism is a good trait. There are compelling authors ought there that tempt you to change investment strategies when you read their book. There needs to be back and forth, peer review, to better refine ideas. I suppose a compelling case can be made for about anything.
A fool and his money are good for business.

User avatar
knpstr
Posts: 1798
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by knpstr » Sun May 14, 2017 10:17 am

Random Walker wrote:BigJon,
I believe Taleb is likewise focused on big events, Black Swans, and that is why he advocates a barbell approach. I could be wrong, but pretty sure he uses lots of T bills and some super risky stuff.

Dave
Yes I think he holds treasuries and I believe he loads up on options as his "risky stuff" when he thinks the time is right. So he has limited, pre-determined downside if he's wrong, with massive upside if he's right.
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

User avatar
tadamsmar
Posts: 7433
Joined: Mon May 07, 2007 12:33 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by tadamsmar » Sun May 14, 2017 12:56 pm

The plot in the OP link (actually the no pay link in the 2nd post on the thread) shows Value doubling the market since 1993 when Fama and French made their case (or since 2000 also, since it lost ground during the dot com boom).

But, what real investment did that? Not the Vanguard Value Index Fund (VIVAX) although it did well vs the market since 1993. Of course that is large cap.

Probably not slice and dice, but I don't have the data.

What could a Boglehead could have implemented back in 1993 or 2000 that would have doubled the market by this point?

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by NiceUnparticularMan » Sun May 14, 2017 1:20 pm

tadamsmar wrote:The plot in the OP link (actually the no pay link in the 2nd post on the thread) shows Value doubling the market since 1993 when Fama and French made their case (or since 2000 also, since it lost ground during the dot com boom).

But, what real investment did that? Not the Vanguard Value Index Fund (VIVAX) although it did well vs the market since 1993. Of course that is large cap.

Probably not slice and dice, but I don't have the data.

What could a Boglehead could have implemented back in 1993 or 2000 that would have doubled the market by this point?
Since 2000, $10,000 in DFA's Large Value fund would get you $40,220, versus $24,188 for VTSMX. DFA's Small Value fund would get you $60,922. Half and half DFA's Large Value and Small Value would get you $50,131.

stlutz
Posts: 4011
Joined: Fri Jan 02, 2009 1:08 am

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by stlutz » Sun May 14, 2017 2:52 pm

The plot in the OP link (actually the no pay link in the 2nd post on the thread) shows Value doubling the market since 1993 when Fama and French made their case (or since 2000 also, since it lost ground during the dot com boom).

But, what real investment did that? Not the Vanguard Value Index Fund (VIVAX) although it did well vs the market since 1993. Of course that is large cap.
Just to provide fuller details:

VG Large Cap since inception (11/1992):

Growth: 9.33%
Value: 9.46

VG Midcap since inception (8/2006):

Growth: 8.52%
Value: 8.78%

VG Smallcap since inception (5/1998):

Growth: 8.38%
Value: 8.78%

Outperformance by value in all 3 cases. Dominating outperformance in line with the chart shown? No.

User avatar
matjen
Posts: 1694
Joined: Sun Nov 20, 2011 11:30 pm

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by matjen » Sun May 14, 2017 3:16 pm

tadamsmar wrote: What could a Boglehead could have implemented back in 1993 or 2000 that would have doubled the market by this point?
iShares S&P Small-Cap 600 Value (IJS) has been around since mid-2000. It has long been spoken favorably of on this board. It has more than doubled the S&P500 since 2000. All these articles people periodically post about anomalies being weak are straw men as far as I am concerned. The academic practitioners like Fama French and Asness really only discuss and incorporate a robust few not 200.


Image
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
grabiner
Advisory Board
Posts: 20839
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by grabiner » Sun May 14, 2017 6:26 pm

matjen wrote:
tadamsmar wrote: What could a Boglehead could have implemented back in 1993 or 2000 that would have doubled the market by this point?
iShares S&P Small-Cap 600 Value (IJS) has been around since mid-2000. It has long been spoken favorably of on this board. It has more than doubled the S&P500 since 2000. All these articles people periodically post about anomalies being weak are straw men as far as I am concerned. The academic practitioners like Fama French and Asness really only discuss and incorporate a robust few not 200.


Image
While I like small-cap value (and overweight it myself), this chart overestimates the advantage of small-cap value, because it has only half of the Internet bubble. Growth stocks outperformed value stocks in 1997-1999, then underperformed when the market fell in 2000-2002. (Still, Vanguard Small-Cap Value Index, which has been around since 1998 and thus did miss both halves of the bubble, is also well ahead of the S&P 500.)
David Grabiner

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: WSJ: Most market anomalies academics have identified don’t exist

Post by NiceUnparticularMan » Sun May 14, 2017 6:37 pm

I believe DFSVX (DFA's Small Value Index fund) is the oldest such fund--it has been around since 1993.

Portfolio Visualizer says that starting in January 1994, $10,000 in DFSVX would get you $133,632, versus $79,293 for VTSMX. That's not quite twice, thanks to the dot-com era--there is about a three-year period from early 1998 to early 2001 where VTSMX is ahead--but it isn't bad.

Post Reply