AQR still singing the same song

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
HippoSir
Posts: 133
Joined: Tue Jul 03, 2018 2:56 pm

Re: AQR still singing the same song

Post by HippoSir »

Ketawa wrote: Thu Sep 17, 2020 2:00 pm Fun to see all the people who endlessly bash factor investing in this thread with the same tired arguments. Meanwhile, I'm happily plugging along with my AQR funds, which I have held since 2014. My entire IRA is AQR funds. QSMLX has outperformed every other noteworthy small cap value fund since its inception, including VBR and IJS: link.
Isn't QSMLX multifactor rather than small-value?

EDIT: Per morningstar it actually has a growth tilt, not a value tilt:

https://www.morningstar.com/funds/xnas/qsmlx/portfolio
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Ketawa wrote: Thu Sep 17, 2020 2:00 pm Fun to see all the people who endlessly bash factor investing in this thread with the same tired arguments. Meanwhile, I'm happily plugging along with my AQR funds, which I have held since 2014. My entire IRA is AQR funds. QSMLX has outperformed every other noteworthy small cap value fund since its inception, including VBR and IJS: link.
Too bad you haven't held it since inception.

You didn't say when in 2014, so I plugged in January 1, 2014

VBR - $10,000 has grown to $13,587
QSMLX - $10,000 has grown to $13,090

Still respectable, but you've gotten no advantage holding an "alt" fund (and paying higher fees).

It is possible to pick dates where QSMLX has done slightly better than VBR... but "out-performed" is a stretch (I guess, technically, one more dollar is out-performed, but usually that term connotates winning by a sizable margin).

They've done about the same. So far, I'm not seeing any special sauce coming from the extra fees you are paying. (Or, there IS special sauce, since AQR is matching the index even with higher fees, but the extra returns just matches the fees, so THEY get the benefit of the secret sauce, not their investors).
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
Topic Author
garlandwhizzer
Posts: 2993
Joined: Fri Aug 06, 2010 3:42 pm

Re: AQR still singing the same song

Post by garlandwhizzer »

Ketawa wrote:

Fun to see all the people who endlessly bash factor investing in this thread with the same tired arguments. Meanwhile, I'm happily plugging along with my AQR funds, which I have held since 2014. My entire IRA is AQR funds. QSMLX has outperformed every other noteworthy small cap value fund since its inception, including VBR and IJS: link.
The problem is that QSMLX has produced a massive opportunity cost to your portfolio relative to VTI (I factor, beta) since 2014. From Portfolio Visualizer:

https://www.portfoliovisualizer.com/bac ... ion2_2=100

A 10K investment in QSMLX would have netted a total positive return $6,811 over that entire 5+ year period. The same 10K position in VTI would have netted over the same time period a $15,156 gain. That is to say thus far on the 10K investment you have paid an opportunity cost of $8345 for your abiding faith in QSMLX. That amounts to more 83% of your entire initial 10K investment. In an age like now when we expect forward equity returns to be substantially lower than historical rates, 83% opportunity cost is a lot pay up front going forward.

But the results of QSMLX were actually worse than that. It massively underperform a single factor fund, VTI (beta ) that has an expense ratio of 0.04%, but in doing so it had a much higher standard deviation, a much worse worst year, a much worse maximum drawdown (36% versus 21%), and consequently a massively lower Sharpe ratio (.44 versus .91). Risk adjusted returns were less than half a cheap single factor fund. I am surprised in view of this track record that Ketawa is still happy plugging along with AQR funds, but I wish him well as I do all investors.

SCV never seems to have any problem finding the increased risk/volatility of small and value. The above demonstrates that clearly. What is has had terrible trouble for more than a decade is finding outperformance. It has instead found massive underperformance. True believers in factor theory stick with it. I am void of true belief and instead a skeptic not just about factor theory but in general. At some point, accumulated opportunity costs dig such a deep hole that even if SCV comes back it may take decades to make up lost ground or never do it. RZV, a deep value microcap fund is already there. This theory is derived from an unrealistic model that inflates positives and ignores negatives, employed in the distant past in totally different macroeconomy (robust growth/bricks and mortar/manufacturing/ industrial economy) plus totally different inflation/interest rate environment (both high when SCV shone). Does it still work in today's information economy that disrupts traditional bricks and mortar plus zero rates/ultra low inflation/decade long stagnant economic growth? interestingly this question does not seem to occur to many factor adherents. I believe that economics rather than factor theory drive which factors thrive and beta has been perfectly positioned for our economic/rate/inflation situation in the last 13 years. Investors have paid up massively for growth because it is so rare. Factor theory may turn out to be a frigid bedfellow if current economic/inflation/interest rate conditions persist. Any some point the hole can get too deep.

One final point. There is great controversy about how best for funds to define and capture the SCV premium. SCV funds choose a wide variety of approaches. It may well turn out that in order to harvest it long term premium you have to pick not just any SCV fund but the right one, since there is a wide dispersion of SCV fund investing results. So you have to pick the right fund, not an easy job with so many out there. There is no such controversy with harvesting beta. You can harvest beta fully at a cost of 0.04%/yr with TSM. Beta is my favorite factor by a long, long shot and it can be reliably harvested for almost nothing. I am very happy that TSM has been by far my largest US. equity holding for the last 15 years. I have no hole to climb out of. If fact my problem is the opposite. I'm on a peak, a point at which I'm considering whether we're currently overpaying too much for growth and the time has come to diversify a bit more.

Garland Whizzer
MotoTrojan
Posts: 10659
Joined: Wed Feb 01, 2017 8:39 pm

Re: AQR still singing the same song

Post by MotoTrojan »

HippoSir wrote: Thu Sep 17, 2020 2:43 pm
Ketawa wrote: Thu Sep 17, 2020 2:00 pm Fun to see all the people who endlessly bash factor investing in this thread with the same tired arguments. Meanwhile, I'm happily plugging along with my AQR funds, which I have held since 2014. My entire IRA is AQR funds. QSMLX has outperformed every other noteworthy small cap value fund since its inception, including VBR and IJS: link.
Isn't QSMLX multifactor rather than small-value?

EDIT: Per morningstar it actually has a growth tilt, not a value tilt:

https://www.morningstar.com/funds/xnas/qsmlx/portfolio
Statistically significant HML loading on QSMLX so not sure what Morningstar is showing you. Looks like since inception it's had less value but way more size exposure than VBR (closer to SLYV in that regard).

https://www.portfoliovisualizer.com/fac ... sion=false
MotoTrojan
Posts: 10659
Joined: Wed Feb 01, 2017 8:39 pm

Re: AQR still singing the same song

Post by MotoTrojan »

garlandwhizzer wrote: Thu Sep 17, 2020 3:23 pm

One final point. There is great controversy about how best for funds to define and capture the SCV premium. SCV funds choose a wide variety of approaches. It may well turn out that in order to harvest it long term premium you have to pick not just any SCV fund but the right one, since there is a wide dispersion of SCV fund investing results. So you have to pick the right fund, not an easy job with so many out there. There is no such controversy with harvesting beta. You can harvest beta fully at a cost of 0.04%/yr with TSM. Beta is my favorite factor by a long, long shot and it can be reliably harvested for almost nothing. I am very happy that TSM has been by far my largest US. equity holding for the last 15 years. I have no hole to climb out of. If fact my problem is the opposite. I'm on a peak, a point at which I'm considering whether we're currently overpaying too much for growth and the time has come to diversify a bit more.

Garland Whizzer
Strongly disagree that which fund you pick impacts your result, what matters is how much factor exposure you use.

Let's say one person uses VBR to tilt and the other RZV; very different funds. If they both tilted 25% they'd get vastly different results, but that is apples-to-oranges. Here you can see that a 25% VBR tilt (rest in VFINX) is equivalent in factor loading to only 12% in RZV. You'll also note that the returns are close, with the RZV variant actually coming out ahead!

https://www.portfoliovisualizer.com/mat ... tion2_1=75
User avatar
nisiprius
Advisory Board
Posts: 41995
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: AQR still singing the same song

Post by nisiprius »

Ketawa wrote: Thu Sep 17, 2020 2:00 pm...QSMLX has outperformed every other noteworthy small cap value fund since its inception, including VBR and IJS
AQR benchmarks it to the Russell 2000 index. Morningstar classifies it as a small-cap blend fund and also benchmarks it to the Russell 2000. For real-world comparisons, it seems to me that the most obvious choices would be the Vanguard Russell 2000 ETF, VTWO and iShares Russell 2000 ETF, IWM.

QSMLX underperformed the benchmark index (orange) and both of those two ETFs (green and yellow) (all three of which are so close to identical that the three lines overlay each other).

If you add the main Vanguard small-cap fund, NAESX, you'll see that it underperformed NAESX too.

QSMLX made money, and gave you whatever small-cap characteristics you were looking for, but it didn't outperform other funds in the same style box, and it didn't hit AQR's chosen benchmark for the fund. And you could easily have hit that benchmark by using index ETFs instead.

Source

Image
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.
wickywack
Posts: 82
Joined: Thu Jun 11, 2015 8:09 am

Re: AQR still singing the same song

Post by wickywack »

HomerJ wrote: Thu Sep 17, 2020 12:38 pm
wickywack wrote: Wed Sep 16, 2020 4:58 pm One thing that fascinates me about QSPIX ( I think AQR's most discussed fund ) is not is recent underperformance, but its forward claims and what ( I think) that actually should mean for portfolio construction if the claims are accepted.

When it came out, the expectation was:
* equity like returns
* half the volatility of equities
* no correlation with equities or bonds

Some fairly smart folks seem confident of the math here.

That would be remarkable. Remarkable enough, that I don't think QSPIX is a diversifier anymore but actual the foundation of a portfolio - i.e., use stocks and bond to diversify from a QSPIX portfolio rather than the other way around (at least for tax advantaged accounts).

In fact, if I throw this at portfolio visualizer and ask it optimize a traditional 60/40 stock bond portfolio + QSPIX for return at a sufficiently high volatility, it suggests ... 100% QSPIX.

Even at lower target volatilities, QSPIX is the biggest suggested slice.

Do folks who are still confident on QSPIX going forward actually allocate that heavily to it? If not, why not?
Why would you accept it's forward claims if it's past claims have been false?
Elysium wrote: Thu Sep 17, 2020 1:38 pm wickywack: Smart people stay away from investments that have proven to be nothing but losing money instead of hoping for something different to happen. There is no need when the market returns have been enough. As John C. Bogle said, the "enemy of a good plan is the pursuit of the perfect plan".
Perhaps I'm being too binary about it. I'd expect folks to either believe in the fund and fully invest or not believe in the fund and not invest altogether (assuming access, tax advantaged account). I don't understand why you'd just add a small allocation though. If you believe in factors / factor diversification and you believe that market is just another factor out of several, then why not an "optimal" allocation? In that case, I think makes sense that you'd have much heavier weighting of QSPIX (with four factors) than a total market fund (with just one). That's what the numbers above appear to suggest.

FWIW, I don't invest in it. I don't have sufficient understanding of why factors (other than market) should persist going forward.
User avatar
Rick Ferri
Posts: 9212
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: AQR still singing the same song

Post by Rick Ferri »

Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
MotoTrojan
Posts: 10659
Joined: Wed Feb 01, 2017 8:39 pm

Re: AQR still singing the same song

Post by MotoTrojan »

Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
This sounds great!!!
User avatar
Robert T
Posts: 2740
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: AQR still singing the same song

Post by Robert T »

.
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Robert T wrote: Fri Sep 18, 2020 9:09 pm .
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
Here's an article from 2018

https://citywireusa.com/professional-bu ... d/a1179846
Starting January 30, 2019, the fund will change its name to the AQR Multi-Asset fund to reflect a shift in investment strategy.

As a result of the changes the fund will be able to gain equity exposure and engage in security selection through specific names as opposed to only through derivatives.

In addition, the fund will have greater flexibility in how it allocates money across various asset classes.

Whereas the fund previously sought ‘to balance the allocation of risk across asset classes,’ it will now allocate assets in a way that ‘avoids excessive risk exposure to any single asset class (e.g., equities, bonds, commodities) or risk premium (e.g., equity risk, duration risk, currency risk),’ according to the filing.
I said, at the time, in another thread...
AQR losing money one year is no big deal. People pulling money out is normal after a bad year. AQR having to lay off people because of all the redemptions is not great, but understandable.

AQR changing their models and implementation (of only one fund so far) is HUGE. That means they don't have faith in at least one of their models anymore, and what they are SELLING is faith in their super-quant super-math super-tested-historical models
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
000
Posts: 2805
Joined: Thu Jul 23, 2020 12:04 am

Re: AQR still singing the same song

Post by 000 »

HomerJ wrote: Fri Sep 18, 2020 11:48 pm
Robert T wrote: Fri Sep 18, 2020 9:09 pm .
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
Here's an article from 2018

https://citywireusa.com/professional-bu ... d/a1179846
Starting January 30, 2019, the fund will change its name to the AQR Multi-Asset fund to reflect a shift in investment strategy.

As a result of the changes the fund will be able to gain equity exposure and engage in security selection through specific names as opposed to only through derivatives.

In addition, the fund will have greater flexibility in how it allocates money across various asset classes.

Whereas the fund previously sought ‘to balance the allocation of risk across asset classes,’ it will now allocate assets in a way that ‘avoids excessive risk exposure to any single asset class (e.g., equities, bonds, commodities) or risk premium (e.g., equity risk, duration risk, currency risk),’ according to the filing.
I said, at the time, in another thread...
AQR losing money one year is no big deal. People pulling money out is normal after a bad year. AQR having to lay off people because of all the redemptions is not great, but understandable.

AQR changing their models and implementation (of only one fund so far) is HUGE. That means they don't have faith in at least one of their models anymore, and what they are SELLING is faith in their super-quant super-math super-tested-historical models
When do you think it is ok to stray from the previously chosen course? Both as an individual investor and as a businessman selling active management?
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

HomerJ wrote: Fri Sep 18, 2020 11:48 pm
Robert T wrote: Fri Sep 18, 2020 9:09 pm .
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
Here's an article from 2018

https://citywireusa.com/professional-bu ... d/a1179846
Starting January 30, 2019, the fund will change its name to the AQR Multi-Asset fund to reflect a shift in investment strategy.

As a result of the changes the fund will be able to gain equity exposure and engage in security selection through specific names as opposed to only through derivatives.

In addition, the fund will have greater flexibility in how it allocates money across various asset classes.

Whereas the fund previously sought ‘to balance the allocation of risk across asset classes,’ it will now allocate assets in a way that ‘avoids excessive risk exposure to any single asset class (e.g., equities, bonds, commodities) or risk premium (e.g., equity risk, duration risk, currency risk),’ according to the filing.
I said, at the time, in another thread...
AQR losing money one year is no big deal. People pulling money out is normal after a bad year. AQR having to lay off people because of all the redemptions is not great, but understandable.

AQR changing their models and implementation (of only one fund so far) is HUGE. That means they don't have faith in at least one of their models anymore, and what they are SELLING is faith in their super-quant super-math super-tested-historical models
Now, you will get the "well, it is important to make changes as new information become available. You would be a fool to ignore new information." argument.

At least that is what i got when i asked about such changes from another favorite prognosticator here. I never did get answer to the question, how do you know when you should "stick to the model" (Very Important because the brilliant strategies can under perform for long periods of time!), or abandon it for the latest shiny object (because you know you would be a fool to not incorporate new information!)

Sigh...
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

000 wrote: Fri Sep 18, 2020 11:53 pm
HomerJ wrote: Fri Sep 18, 2020 11:48 pm
Robert T wrote: Fri Sep 18, 2020 9:09 pm .
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
Here's an article from 2018

https://citywireusa.com/professional-bu ... d/a1179846
Starting January 30, 2019, the fund will change its name to the AQR Multi-Asset fund to reflect a shift in investment strategy.

As a result of the changes the fund will be able to gain equity exposure and engage in security selection through specific names as opposed to only through derivatives.

In addition, the fund will have greater flexibility in how it allocates money across various asset classes.

Whereas the fund previously sought ‘to balance the allocation of risk across asset classes,’ it will now allocate assets in a way that ‘avoids excessive risk exposure to any single asset class (e.g., equities, bonds, commodities) or risk premium (e.g., equity risk, duration risk, currency risk),’ according to the filing.
I said, at the time, in another thread...
AQR losing money one year is no big deal. People pulling money out is normal after a bad year. AQR having to lay off people because of all the redemptions is not great, but understandable.

AQR changing their models and implementation (of only one fund so far) is HUGE. That means they don't have faith in at least one of their models anymore, and what they are SELLING is faith in their super-quant super-math super-tested-historical models
When do you think it is ok to stray from the previously chosen course? Both as an individual investor and as a businessman selling active management?
Doesn't matter what I think... Asness said in 2015 interview that sticking with the model is important.

I'm just pointing out that in 2019, they didn't stick with the model with at least one of their funds. I haven't followed AQR closely enough to know if they have made changes in other funds as well.

I like his time dilation quote... I'm guessing the past 3-5 years have felt like 20 to him.

But the billions he's collected in fees probably helps soothe the pain.
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

HomerJ wrote: Sat Sep 19, 2020 12:58 am
000 wrote: Fri Sep 18, 2020 11:53 pm
HomerJ wrote: Fri Sep 18, 2020 11:48 pm
Robert T wrote: Fri Sep 18, 2020 9:09 pm .
Here are some extras from a Q&A with Asness from Lasse Pedersen’s book Efficiently Inefficient (2015).

My underlines.
Well, the single biggest difference between real world and academia is— this sounds over scientific— time dilation. I’ll explain what I mean. This is not relativistic time dilation as the only time I move at speeds near light is when there is pizza involved. But to borrow the term, your sense of time does change when you are running real money. Suppose you look at a cumulative return of a strategy with a Sharpe ratio of 0.7 and see a three-year period with poor performance. It does not faze you one drop. You go: “Oh, look, that happened in 1973, but it came back by 1976, and that’s what a 0.7 Sharpe ratio does.” But living through those periods takes— subjectively, and in wear and tear on your internal organs— many times the actual time it really lasts. If you have a three-year period where something doesn’t work, it ages you a decade. You face an immense pressure to change your models, you have bosses or clients who lose faith, and I cannot explain the amount of discipline that you need.
Discipline is important. We do not think we’re more immune to psychological biases than others, but following the models helps. If we followed them with less discipline, we run the serious risk of reintroducing the exact biases we are trying to exploit! For instance, if people run from stocks with any problems, making value stocks too cheap and attractive buys long-term, if we use our judgment to selectively override our models, perhaps we undo precisely the bet we want to make, in order to make ourselves more “comfortable”? Discipline is not always easy, by the way. It’s really hard to stick to a strategy. But when people cave and disregard their models, they seem to usually do this within an hour and a half of the worst possible time to cave. Admittedly, that’s not a quant study, but it’s been my experience. The difficulty of sticking to the models is part of why they work.
.
Here's an article from 2018

https://citywireusa.com/professional-bu ... d/a1179846
Starting January 30, 2019, the fund will change its name to the AQR Multi-Asset fund to reflect a shift in investment strategy.

As a result of the changes the fund will be able to gain equity exposure and engage in security selection through specific names as opposed to only through derivatives.

In addition, the fund will have greater flexibility in how it allocates money across various asset classes.

Whereas the fund previously sought ‘to balance the allocation of risk across asset classes,’ it will now allocate assets in a way that ‘avoids excessive risk exposure to any single asset class (e.g., equities, bonds, commodities) or risk premium (e.g., equity risk, duration risk, currency risk),’ according to the filing.
I said, at the time, in another thread...
AQR losing money one year is no big deal. People pulling money out is normal after a bad year. AQR having to lay off people because of all the redemptions is not great, but understandable.

AQR changing their models and implementation (of only one fund so far) is HUGE. That means they don't have faith in at least one of their models anymore, and what they are SELLING is faith in their super-quant super-math super-tested-historical models
When do you think it is ok to stray from the previously chosen course? Both as an individual investor and as a businessman selling active management?
Doesn't matter what I think... Asness said in 2015 interview that sticking with the model is important.

I'm just pointing out that in 2019, they didn't stick with the model with at least one of their funds. I haven't followed AQR closely enough to know if they have made changes in other funds as well.

I like his time dilation quote... I'm guessing the past 3-5 years have felt like 20 to him.

But the billions he's collected in fees probably helps soothe the pain.
I don't understand all the hand wringing about how Asness has not met the goals with his funds. It seems to be a miss-reading of what the actual goals are of the funds. As you point, he seems to have been fabulously successful in achieving his goals.
Once in a while you get shown the light, in the strangest of places if you look at it right.
000
Posts: 2805
Joined: Thu Jul 23, 2020 12:04 am

Re: AQR still singing the same song

Post by 000 »

HomerJ wrote: Sat Sep 19, 2020 12:58 am Doesn't matter what I think... Asness said in 2015 interview that sticking with the model is important.

I'm just pointing out that in 2019, they didn't stick with the model with at least one of their funds. I haven't followed AQR closely enough to know if they have made changes in other funds as well.

I like his time dilation quote... I'm guessing the past 3-5 years have felt like 20 to him.

But the billions he's collected in fees probably helps soothe the pain.
Gotcha, thanks.

It seems that the apparent inability of active managers to stay the course is all the more reason to not pay them, i.e. either buy the index or do your own active management.
User avatar
Robert T
Posts: 2740
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: AQR still singing the same song

Post by Robert T »

.
I think Asness/AQR has been fairly consistent on value and momentum (initial focus), with quality added later.

Re: Interaction of Value and Momentum Strategies and Value and Momentum Everywhere

Its earliest long-only funds focused (and still focus) on value and momentum (International equity, global equity, and momentum only funds), with quality added (multi-style, and defensive equity - the latter combining quality and low volatility). Even some of the long-short funds (equity market neutral, long-short equity) include value, momentum, quality screens, with more recent products also adding carry (style premia), and trend (alternative risk premia) to value, momentum, and quality screens.

So recent underperformance of value cuts across many (perhaps most), but not all, of the AQR funds.

My concern in adding more ‘factors’ to stock screens is: (i) dilution, if a large sample of stocks is retained; or (ii) substantial reductions in the number of securities meeting all criteria which introduces more idiosyncratic risk. As Fama said earlier there are significant diminishing returns to adding additional factors.e.g. From 1975-2012 adding profitability screens to DFA’s Targeted Value series resulted in a 0.2% higher annualized return!, same effect for DFA Vector. Yet there was (and still is by some) a huge DFA marketing spin on it.

And I also think the component versus integrated approach is oversold. Again, for the ‘rational’ voice on this I turn to Bernstein – “tilt is tilt, no matter where it comes from.” If you have a specific factor load target, then there is very little/no difference on whether this is achieved through a component or integrated approach. Choose the lowest cost option.

Fees at AQR are relatively high, but would note that in July, they reduced fees on 8 of the funds. Perhaps competitive forces at work (rather that any altruist feeling).

I still get a bit of a ‘black-box’ feel with AQR. For example, the longer-term back-tested data on their long only funds are not available (as they are for MSCI, RAFI, CRSP etc), so much harder to understand the likely long-term characteristics of the funds. The exception is the momentum series which are available, and allow for a better and more informed comparison with FF momentum, and MSCI momentum. Based on this my sense was MSCI was better (less negative alpha in the back-tested data). For the other AQR funds without available long-term back-tested data – the premise seems to be ‘trust us’.

Personally, I have actually learned a lot from Asness, and AQR has some interesting products (even though I don’t use any of them). Time will tell whether Asness (alone) simply gets richer, or whether AQR clients will too relative to 'market benchmarks' (it has certainly not been a proportionate matching so far). Ironically, going forward this seems to rely a lot on 'value'.
.
User avatar
nisiprius
Advisory Board
Posts: 41995
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: AQR still singing the same song

Post by nisiprius »

Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Last edited by nisiprius on Sat Sep 19, 2020 12:08 pm, edited 1 time 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
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 12:19 pm
nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 12:46 pm
Steve Reading wrote: Sat Sep 19, 2020 12:19 pm
nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 2:07 pm
marcopolo wrote: Sat Sep 19, 2020 12:46 pm
Steve Reading wrote: Sat Sep 19, 2020 12:19 pm
nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
To be honest, given your example, the original "vegetarianism is healthier" was a disturbingly poor conclusion. Any study that did not control for these other things and came to the conclusion that vegetarianism was healthier should be retracted. To claim you are just saying the same thing you have always said IS revisionist history.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Steve Reading wrote: Sat Sep 19, 2020 2:07 pm
marcopolo wrote: Sat Sep 19, 2020 12:46 pm
Steve Reading wrote: Sat Sep 19, 2020 12:19 pm
nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular
Does size matter? Certainly in financial markets, many researchers have questioned whether it does. In a new paper, we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.
Cliff Asness, September 18, 2020: There Is No Size Effect
So, net of using the more accurate databases today versus those used in the original work and this adjustment for underestimated betas (using lags), you really don’t find any size effect. Zippo. Nada.
I am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
That's a good analogy.. but it proves that Asness made a poor conclusion 5 years ago, and now he's refining it.

The problem is, what if there are MORE variables he hasn't accounted for, and 5 years from now, he has to refine the model again?

Economics has too many unknown variables. No one knows enough to accurately predict the odds or even fully understand the game.
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
Elysium
Posts: 3221
Joined: Mon Apr 02, 2007 6:22 pm

Re: AQR still singing the same song

Post by Elysium »

Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
Rick, This is a great idea. I think Mr.Asness should begin with an apology to all investors in his funds who have lost so much of asset value over the years after they placed trust in his abilities to provide a decent risk adjusted return. Since I don't think that will be happening, I will have no questions.
Last edited by Elysium on Sat Sep 19, 2020 2:58 pm, edited 1 time in total.
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 2:17 pm
Steve Reading wrote: Sat Sep 19, 2020 2:07 pm
marcopolo wrote: Sat Sep 19, 2020 12:46 pm
Steve Reading wrote: Sat Sep 19, 2020 12:19 pm
nisiprius wrote: Sat Sep 19, 2020 11:56 am Cliff Asness, January 23rd, 2015: The Small-Firm Effect is Real, and It's Spectacular

Cliff Asness, September 18, 2020: There Is No Size EffectI am sure it is possible to reconcile these points of view, but it is easier for me to just not get confused in the first place, then to get confused in 2015 and straightened out five years later.
Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:
In our size/junk paper we went and confused some people (we really tried not to!). We confirmed the lack of existence of a simple size effect but showed that small stocks were far lower “quality” than big stocks. Thus, any return they generate is, in fact, more impressive as historically quality is a rewarded factor, so small stocks face a rather severe headwind. In English, it is quite a feat for small stocks to even keep up with large stocks (after market beta adjustment) given that small stocks are far lower quality. We found this decidedly non-simple size effect to be impressive. Some have, quite mistakenly and to our great surprise, taken this as a contradiction: “I mean, in one paper you say size doesn’t work but in another you say it does!” Nope, not what we said. In both papers we say that size doesn’t really work alone (the “simple” size effect adjusting for only market beta) and in both papers we say (the main focus of one paper and a section of the other) that only when considering they’re fighting the successful quality factor are small vs. large returns impressive. These are not at all contradictory.
Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
To be honest, given your example, the original "vegetarianism is healthier" was a disturbingly poor conclusion. Any study that did not control for these other things and came to the conclusion that vegetarianism was healthier should be retracted. To claim you are just saying the same thing you have always said IS revisionist history.
Revisionist history to me means that you claimed something and then claimed something different or inconsistent later on. Like how the world was thought to be flat thousands of years ago, but now we know it's round. There was a clear revision here because the world cannot be both flat and round.

The above AQR and vegetarian examples aren't revisionist history because they're not inconsistent, or wrong. They're all facts and continue to be facts.
It is a fact that vegetarians are healthier than non-vegetarians. Not a "poor conclusion", it's the truth. This statement is NOT saying vegetarians are healthier BECAUSE of the diet. So it will be consistent and reasonable to then also claim vegetarianism does not cause additional health benefits over, say, exercise.

Call it "misleading" or "confusing". AQR certainly agrees they confused many. But it's not inconsistent. They don't have to retract any of their papers. They're all perfectly factual and true.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Elysium wrote: Sat Sep 19, 2020 2:55 pm
Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
Rick, This is a great idea. I think Mr.Asness should begin with an apology to all investors in his funds who have lost so much of asset value over the years after they placed trust in his abilities to provide them with a decent risk adjusted return. Since I don't think that will be happening, I will have no questions.
I only have one question, and I seriously would like you to ask it.

"Where are the customer's yachts?"
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
Elysium
Posts: 3221
Joined: Mon Apr 02, 2007 6:22 pm

Re: AQR still singing the same song

Post by Elysium »

HomerJ wrote: Sat Sep 19, 2020 2:58 pm
Elysium wrote: Sat Sep 19, 2020 2:55 pm
Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
Rick, This is a great idea. I think Mr.Asness should begin with an apology to all investors in his funds who have lost so much of asset value over the years after they placed trust in his abilities to provide them with a decent risk adjusted return. Since I don't think that will be happening, I will have no questions.
I only have one question, and I seriously would like you to ask it.

"Where are the customer's yachts?"
Well, that might make him think Bogleheads are anti-capitalists which we aren't :wink: I'm okay him making a decent profit, so long as he is able to provide the investors with what he promised. I don't doubt his intentions or his conviction in the strategy, but at some point you have to realize you have failed the customers and you had it wrong, so an apology is on order.
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

HomerJ wrote: Sat Sep 19, 2020 2:53 pm That's a good analogy.. but it proves that Asness made a poor conclusion 5 years ago, and now he's refining it.

The problem is, what if there are MORE variables he hasn't accounted for, and 5 years from now, he has to refine the model again?
No, Asness' conclusion that "size exists once you control for junk" 5 years ago continues to be true. And nothing in this latest paper refutes or otherwise goes against it. I mean did you even read it? :confused

If at some point Asness found that the size effect did NOT exist any more when controlling for beta and quality, THEN you can say the conclusion 5 years ago is incorrect or poor.

AQR will definitely change its stance and refine its models as they continue to research and learn more. They probably do that constantly. More variables will almost certainly pop up in the future, which will change their investment strategy. They're called Applied Quantitative Research after all. If that makes you uncomfortable, then don't invest :beer
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 2:56 pm
marcopolo wrote: Sat Sep 19, 2020 2:17 pm
Steve Reading wrote: Sat Sep 19, 2020 2:07 pm
marcopolo wrote: Sat Sep 19, 2020 12:46 pm
Steve Reading wrote: Sat Sep 19, 2020 12:19 pm

Not only is it possible to reconcile them. But Cliff himself clears up your confusion on the very next paragraph:



Bolded emphasis is mine.
I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
To be honest, given your example, the original "vegetarianism is healthier" was a disturbingly poor conclusion. Any study that did not control for these other things and came to the conclusion that vegetarianism was healthier should be retracted. To claim you are just saying the same thing you have always said IS revisionist history.
Revisionist history to me means that you claimed something and then claimed something different or inconsistent later on. Like how the world was thought to be flat thousands of years ago, but now we know it's round. There was a clear revision here because the world cannot be both flat and round.

The above AQR and vegetarian examples aren't revisionist history because they're not inconsistent, or wrong. They're all facts and continue to be facts.
It is a fact that vegetarians are healthier than non-vegetarians. Not a "poor conclusion", it's the truth. This statement is NOT saying vegetarians are healthier BECAUSE of the diet. So it will be consistent and reasonable to then also claim vegetarianism does not cause additional health benefits over, say, exercise.

Call it "misleading" or "confusing". AQR certainly agrees they confused many. But it's not inconsistent. They don't have to retract any of their papers. They're all perfectly factual and true.
I think we have different understanding of revisionist history.

in your example of the flat earth, it would be claiming the earth is flat, then discovering it is round and saying "we always said it was round, so no inconsistencies here". I.e., they are revising what they historically said to make it look like they were always right.

In the original paper, they sure did make it seem like size was independently useful. Now they discover it is not, and they are going back and saying "no inconsistency" as if they always said (or intended to say) it is just coincidence when looked at in isolation to other causes.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 3:14 pm
Steve Reading wrote: Sat Sep 19, 2020 2:56 pm
marcopolo wrote: Sat Sep 19, 2020 2:17 pm
Steve Reading wrote: Sat Sep 19, 2020 2:07 pm
marcopolo wrote: Sat Sep 19, 2020 12:46 pm

I don't know for sure, but it sounds a bit like revisiionist history to me. when he said in the older paper:

we resurrect the size premium and restore it to its proper place alongside such stalwarts as value and momentum in terms of its efficacy and robustness.


That sure sounds like he was saying size works (it's a stalwart!). Are you suggesting he is actually saying value and momentum also don't work? Since he equated size to them.
When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
To be honest, given your example, the original "vegetarianism is healthier" was a disturbingly poor conclusion. Any study that did not control for these other things and came to the conclusion that vegetarianism was healthier should be retracted. To claim you are just saying the same thing you have always said IS revisionist history.
Revisionist history to me means that you claimed something and then claimed something different or inconsistent later on. Like how the world was thought to be flat thousands of years ago, but now we know it's round. There was a clear revision here because the world cannot be both flat and round.

The above AQR and vegetarian examples aren't revisionist history because they're not inconsistent, or wrong. They're all facts and continue to be facts.
It is a fact that vegetarians are healthier than non-vegetarians. Not a "poor conclusion", it's the truth. This statement is NOT saying vegetarians are healthier BECAUSE of the diet. So it will be consistent and reasonable to then also claim vegetarianism does not cause additional health benefits over, say, exercise.

Call it "misleading" or "confusing". AQR certainly agrees they confused many. But it's not inconsistent. They don't have to retract any of their papers. They're all perfectly factual and true.
I think we have different understanding of revisionist history.

in your example of the flat earth, it would be claiming the earth is flat, then discovering it is round and saying "we always said it was round, so no inconsistencies here". I.e., they are revising what they historically said to make it look like they were always right.

In the original paper, they sure did make it seem like size was independently useful. Now they discover it is not, and they are going back and saying "no inconsistency" as if they always said (or intended to say) it is just coincidence when looked at in isolation to other causes.
Haha all right man. Sure.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 3:30 pm
marcopolo wrote: Sat Sep 19, 2020 3:14 pm
Steve Reading wrote: Sat Sep 19, 2020 2:56 pm
marcopolo wrote: Sat Sep 19, 2020 2:17 pm
Steve Reading wrote: Sat Sep 19, 2020 2:07 pm

When you look at size with no other factor, size worked. When you control for beta, it didn't. When you control for beta and quality, it worked again.

It's like this:
Doctor = Vegetarians are statistically healthier than non-vegetarians.
Also the same Doctor = When you compare populations that exercise as much as vegetarians, control their calories like vegetarians, are as health-minded, etc, then you don't actually see the vegetarian population being healthier. These were confounding variables.
You = Idk Doc. Sounds to me like revisionist history to me. Before you said vegetarianism was healthier, then you said it doesn't have any benefits.

^Depending on what you control for, things "work" or "don't work". There's no AQR contradiction here.
To be honest, given your example, the original "vegetarianism is healthier" was a disturbingly poor conclusion. Any study that did not control for these other things and came to the conclusion that vegetarianism was healthier should be retracted. To claim you are just saying the same thing you have always said IS revisionist history.
Revisionist history to me means that you claimed something and then claimed something different or inconsistent later on. Like how the world was thought to be flat thousands of years ago, but now we know it's round. There was a clear revision here because the world cannot be both flat and round.

The above AQR and vegetarian examples aren't revisionist history because they're not inconsistent, or wrong. They're all facts and continue to be facts.
It is a fact that vegetarians are healthier than non-vegetarians. Not a "poor conclusion", it's the truth. This statement is NOT saying vegetarians are healthier BECAUSE of the diet. So it will be consistent and reasonable to then also claim vegetarianism does not cause additional health benefits over, say, exercise.

Call it "misleading" or "confusing". AQR certainly agrees they confused many. But it's not inconsistent. They don't have to retract any of their papers. They're all perfectly factual and true.
I think we have different understanding of revisionist history.

in your example of the flat earth, it would be claiming the earth is flat, then discovering it is round and saying "we always said it was round, so no inconsistencies here". I.e., they are revising what they historically said to make it look like they were always right.

In the original paper, they sure did make it seem like size was independently useful. Now they discover it is not, and they are going back and saying "no inconsistency" as if they always said (or intended to say) it is just coincidence when looked at in isolation to other causes.
Haha all right man. Sure.
Curious what you find funny.

I am trying to understand your position.

If someone did a study and came to the conclusion that "poor people are less intelligent" without controlling for lack of educational opportunities, access to nutritional foods, etc.

Then 5 years later looked into those items and discovered that when correcting for those things there was no link between being poor and performance on intellignece tests.

You think their initial conclusions are still correct, and consistent? And they can legitimately claim what they were saying was correct all along?

If so, then you are right AQR is consistent.

I disagree, but we are each allowed our opinions.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Dead Man Walking
Posts: 1001
Joined: Wed Nov 07, 2007 6:51 pm

Re: AQR still singing the same song

Post by Dead Man Walking »

Having read this thread, I am leaning towards the double negative that usually appears in these threads: Nobody knows nothing.

DMW
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 3:42 pm If someone did a study and came to the conclusion that "poor people are less intelligent" without controlling for lack of educational opportunities, access to nutritional foods, etc.

Then 5 years later looked into those items and discovered that when correcting for those things there was no link between being poor and performance on intellignece tests.

You think their initial conclusions are still correct, and consistent? And they can legitimately claim what they were saying was correct all along?
No no, this is completely different, I see your confusion. It's not that AQR found out something, then learned more, and revised their previous stance. AQR has since day 1 said the same message. It's just that in the same paper, AQR said comments like "size doesn't work past beta and liquidity" and also said "but it does work with beta, liquidity and quality".

It's like the Doctor telling you "hey vegetarianism is associated with higher health. But it's actually because of more exercise". And then you'd respond "well, which is Doctor, does vegetarianism give more health or not? You sound very inconsistent".

Or using your example above, it'd be like a study saying "we find that poor people are less intelligent. But once we control for, say, country, we find that's not the case at all".

You're saying that's inconsistent. I'm saying it's not at all. But if you did take sentences out of context from that study, you might say "wait, they're saying poor people are less intelligent AND poor people aren't less intelligent".
marcopolo wrote: Sat Sep 19, 2020 3:42 pm If so, then you are right AQR is consistent.

I disagree, but we are each allowed our opinions.
Except AQR never even flip-flopped the way you describe above. Since the beginning, AQR has consistently said size works by itself, doesn't when you control for the confounding variables beta and liquidity, and does again once you add profitability on top. See for yourself:

1) Jan 2015: The Small-Firm Effect Is Real, and It's Spectacular
These assaults have rendered the pure small-cap premium as marginally significant, at best, and “not real” at worst.
...
The many formidable challenges to size all disappear with the introduction of the QMJ factor. It turns out that small stocks are quite “junky” versus their larger counterparts, and it is this exposure that is dampening their performance. Once you account for the QMJ exposure a large and significant size premium re-emerges.
TLDR: "FF size is very flimsy/weak once you control for beta and liquidity. But it's very robust if you add profitability"
2) May 2015: Size Matters If You Control Your Junk
The size premium has been accused of having a weak historical record, being meager relative to other factors, varying significantly over time, weakening after its discovery, being concentrated among microcap stocks, residing predominantly in January, relying on price-based measures, and being weak internationally. We find, however, that these challenges disappear when controlling for the quality, or its inverse, junk, of a firm. A significant size premium emerges...
TLDR: "FF size is very flimsy/weak once you control for beta and liquidity. But it's very robust if you add profitability"
3) May 2018: Fact, Fiction, and the Size Effect
FICTION: THE SIZE EFFECT IS LIKELY MORE THAN JUST A LIQUIDITY EFFECT
...
FICTION: THE SIZE EFFECT WORKS IN OTHER EQUITY MARKETS
...
Why does the size effect become significantly stronger when controlling for the profitability factor? Because, as Asness et al. (2018) showed, the size effect is confounded by a very powerful quality versus junk effect. They investigated the relationship between size and quality and found that controlling for quality not only resurrects the size premium and elevates it significantly but also helps resolve some of the aforementioned patterns associated with size.
TLDR: "FF size is very flimsy/weak once you control for beta and liquidity. But it's very robust if you add profitability"
3) September 2020: There Is No Size Effect: Daily Edition
We, consistently across both papers, find the simple small firm effect doesn’t exist, as small firms do not historically defeat large ones by more than their market beta. And we, consistently across both papers, find that small firms indeed look much more impressive than large ones if you also adjust for their lower quality (not generally what people have called the small firm effect!). The message is consistent and straightforward (if indeed more complicated than when quality is left out).
The problem is that people read the Fact and Fiction paper, read how they keep saying FF size doesn't actually work past beta but then claim it works (when added to quality) and they're like "wait OMG contradiction".

And there's no contradiction at all. None. Zero. Nada. Seriously, what do you want from these guys? :confused
Last edited by Steve Reading on Sat Sep 19, 2020 5:52 pm, edited 1 time in total.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 3:42 pm Curious what you find funny.
I laugh by the way because it's evident you haven't read these papers yet have plenty of opinions.
marcopolo wrote: Sat Sep 19, 2020 3:14 pm In the original paper, they sure did make it seem like size was independently useful. Now they discover it is not, and they are going back and saying "no inconsistency" as if they always said (or intended to say) it is just coincidence when looked at in isolation to other causes.
I will set the record straight one more time. AQR believed, believes and has always believed:
1) Small caps beat large caps historically, i.e size works if no factors are added.
2) Small caps don't beat large caps once you control for beta/liquidity confounding variables, i.e. Fama-French Size doesn't actually work (since that's in the 3-Factor world).
3) Small caps do beat large caps when you control for beta/liquidity, if you also control for quality, i.e. the size approach of AQR has worked historically.

This theory of yours that AQR though "size worked", but then discovered it didn't, so then scrambled to find a way to "revive it" to save face is just straight false.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 5:28 pm
marcopolo wrote: Sat Sep 19, 2020 3:42 pm Curious what you find funny.
I laugh by the way because it's evident you haven't read these papers yet have plenty of opinions.
marcopolo wrote: Sat Sep 19, 2020 3:14 pm In the original paper, they sure did make it seem like size was independently useful. Now they discover it is not, and they are going back and saying "no inconsistency" as if they always said (or intended to say) it is just coincidence when looked at in isolation to other causes.
I will set the record straight one more time. AQR believed, believes and has always believed:
1) Small caps beat large caps historically, i.e size works if no factors are added.
2) Small caps don't beat large caps once you control for beta/liquidity confounding variables, i.e. Fama-French Size doesn't actually work (since that's in the 3-Factor world).
3) Small caps do beat large caps when you control for beta/liquidity, if you also control for quality, i.e. the size approach of AQR has worked historically.

This theory of yours that AQR though "size worked", but then discovered it didn't, so then scrambled to find a way to "revive it" to save face is just straight false.

I have read both the posts on his blog. Admittedly, have not read the source papers, but I assume he is representing what is in the paper accurately in his blog.

Perhaps, the reason we came away with different interpretations of he had to say is at least somewhat based on prior feelings on the topic at hand.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Elysium
Posts: 3221
Joined: Mon Apr 02, 2007 6:22 pm

Re: AQR still singing the same song

Post by Elysium »

Steve Reading wrote: Sat Sep 19, 2020 3:10 pm AQR will definitely change its stance and refine its models as they continue to research and learn more. They probably do that constantly. More variables will almost certainly pop up in the future, which will change their investment strategy. They're called Applied Quantitative Research after all. If that makes you uncomfortable, then don't invest :beer
Yes, they will keep experimenting using other people's money, with products and models that were faulty to begin with, that is the meaning of above statement. While they keep getting fat off the fees.
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Steve Reading wrote: Sat Sep 19, 2020 3:10 pm AQR will definitely change its stance and refine its models as they continue to research and learn more. They probably do that constantly. More variables will almost certainly pop up in the future, which will change their investment strategy. They're called Applied Quantitative Research after all. If that makes you uncomfortable, then don't invest :beer
LOL... Of course, that makes me uncomfortable. So I don't invest. Total Stock Market Index Fund isn't changing its formula every few weeks.

I'm amazed that you are not uncomfortable after losing all that money.

If one had serious money invested in AQR, and one had lost 30% since 2015 while the boring old stock index funds are up 82%, I'm guessing most people would be uncomfortable.

Maybe one had $10,000 invested, it's not a big deal.

If one was older and had $300,000 invested in Cliff's theories, and now they have $200,000 instead of $550,000, I'm thinking missing out on $350,000 over the past five years would make most people question good old Cliff's models that he built in 2015 (and apparently in 2016, 2017, 2018, etc.)

Edit: Also, Cliff says that sticking to the model is important, but you say they change the model continuously. There seems to be a disconnect there.
Last edited by HomerJ on Sat Sep 19, 2020 9:55 pm, edited 3 times in total.
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

HomerJ wrote: Sat Sep 19, 2020 9:12 pm
Steve Reading wrote: Sat Sep 19, 2020 3:10 pm AQR will definitely change its stance and refine its models as they continue to research and learn more. They probably do that constantly. More variables will almost certainly pop up in the future, which will change their investment strategy. They're called Applied Quantitative Research after all. If that makes you uncomfortable, then don't invest :beer
LOL... Of course, that makes me uncomfortable. So I don't invest. Total Stock Market Index Fund isn't changing its formula every few weeks.

I'm amazed that you are not uncomfortable after losing all that money.

I guess you don't have much invested yet. Because if you had serious money invested in AQR, and you had lost 30% since 2015 while the boring old stock index funds are up 82%, I'm guessing most people would be uncomfortable.

Maybe you have $10,000 invested, and it's not a big deal.

If you were older and had $300,000 invested in Cliff's theories, and now you have $200,000 instead of $550,000, I'm thinking missing out on $350,000 over the past five years would make you question old Cliff more.
Huh? :confused

I don’t even use AQR funds. I happen to think the fees are too high for what they offer. I don’t even know what makes you think I use them.

I guess this topic is so polarized that if you try to be objective and fair to AQR, AQR haters immediately think you must invest with them.

FWIW, there’s plenty to dislike about AQR and AQR funds. But this whole “size/quality/beta” argument Marco and you keep pounding about just isn’t one of them. As much as you want it to be, it just isn’t.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

HomerJ wrote: Sat Sep 19, 2020 9:12 pm Edit: Also, Cliff says that sticking to the model is important, but you say they change the model continuously. There seems to be a disconnect there.
No disconnect. AQR doesn’t change strategies based on underperformance. Just because value has done poorly doesn’t mean AQR will stop using value. They know factors under perform. So sticking to the strategy is paramount. Cliff is a huge believer in staying the course, that’s why Bogle and him agreed so much in that one interview they had together.

What does change, probably frequently, is exactly how strategies are implemented. Whether they use risk-parity, or how they define their proprietary value measures. Whether they should use low-volatility only with the quality stocks or with all of the stocks in the fund, etc etc.

So the details and models change but the big picture and general tropes have been the same for nearly 20 years and probably will continue too.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
marcopolo
Posts: 3491
Joined: Sat Dec 03, 2016 10:22 am

Re: AQR still singing the same song

Post by marcopolo »

Steve Reading wrote: Sat Sep 19, 2020 10:02 pm
HomerJ wrote: Sat Sep 19, 2020 9:12 pm
Steve Reading wrote: Sat Sep 19, 2020 3:10 pm AQR will definitely change its stance and refine its models as they continue to research and learn more. They probably do that constantly. More variables will almost certainly pop up in the future, which will change their investment strategy. They're called Applied Quantitative Research after all. If that makes you uncomfortable, then don't invest :beer
LOL... Of course, that makes me uncomfortable. So I don't invest. Total Stock Market Index Fund isn't changing its formula every few weeks.

I'm amazed that you are not uncomfortable after losing all that money.

I guess you don't have much invested yet. Because if you had serious money invested in AQR, and you had lost 30% since 2015 while the boring old stock index funds are up 82%, I'm guessing most people would be uncomfortable.

Maybe you have $10,000 invested, and it's not a big deal.

If you were older and had $300,000 invested in Cliff's theories, and now you have $200,000 instead of $550,000, I'm thinking missing out on $350,000 over the past five years would make you question old Cliff more.
Huh? :confused

I don’t even use AQR funds. I happen to think the fees are too high for what they offer. I don’t even know what makes you think I use them.

I guess this topic is so polarized that if you try to be objective and fair to AQR, AQR haters immediately think you must invest with them.

FWIW, there’s plenty to dislike about AQR and AQR funds. But this whole “size/quality/beta” argument Marco and you keep pounding about just isn’t one of them. As much as you want it to be, it just isn’t.
It is really not any specific strategy that i am questioning. It is the entire notion that really smart people can somehow mine historical data and develop an complex approach that will consistently beat their benchmarks going forward. There is an endless string of individuals and funds that claim to do that, most end up failing. It is especially egregious when they add insult to injury by charging a small fortune for the privilege of investing with them. I also find it interesting that many of the proponents of such complex strategies have convinced themselves that they are so clever, that they feel they need to belittle and question the intelligence of anyone that is not smart enough to grasp just how special their particular complex strategy happens to be.
Once in a while you get shown the light, in the strangest of places if you look at it right.
000
Posts: 2805
Joined: Thu Jul 23, 2020 12:04 am

Re: AQR still singing the same song

Post by 000 »

marcopolo wrote: Sat Sep 19, 2020 10:32 pm It is really not any specific strategy that i am questioning. It is the entire notion that really smart people can somehow mine historical data and develop an complex approach that will consistently beat their benchmarks going forward. There is an endless string of individuals and funds that claim to do that, most end up failing. It is especially egregious when they add insult to injury by charging a small fortune for the privilege of investing with them. I also find it interesting that many of the proponents of such complex strategies have convinced themselves that they are so clever, that they feel they need to belittle and question the intelligence of anyone that is not smart enough to grasp just how special their particular complex strategy happens to be.
+1
User avatar
Steve Reading
Posts: 2460
Joined: Fri Nov 16, 2018 10:20 pm

Re: AQR still singing the same song

Post by Steve Reading »

marcopolo wrote: Sat Sep 19, 2020 10:32 pm It is really not any specific strategy that i am questioning. It is the entire notion that really smart people can somehow mine historical data and develop an complex approach that will consistently beat their benchmarks going forward. There is an endless string of individuals and funds that claim to do that, most end up failing. It is especially egregious when they add insult to injury by charging a small fortune for the privilege of investing with them.
IMO, every single one of those points are reasonable arguments against AQR. I myself agree with them to various extents. They also have been discussed extensively in this forum already so not much point just repeating them (although as you can see, some other posters just love to repeat them).

I just don't think we need to manufacture further evidence against them. In the topic of size, I think AQR has been very evidence-based and consistent in their message. So I'm bothered if others start claiming they aren't consistent, or trying trying to revise statements. Let's be objective about it instead.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
HomerJ
Posts: 15190
Joined: Fri Jun 06, 2008 12:50 pm

Re: AQR still singing the same song

Post by HomerJ »

Steve Reading wrote: Sat Sep 19, 2020 10:17 pmCliff is a huge believer in staying the course, that’s why Bogle and him agreed so much in that one interview they had together.

What does change, probably frequently, is exactly how strategies are implemented. Whether they use risk-parity, or how they define their proprietary value measures. Whether they should use low-volatility only with the quality stocks or with all of the stocks in the fund, etc etc.
None of that appears to me to be "staying the course". I'm afraid we're speaking different languages here.
A Goldman Sachs associate provided a variety of detailed explanations, but then offered a caveat, “If I’m being dead-### honest, though, nobody knows what’s really going on.”
langlands
Posts: 572
Joined: Wed Apr 03, 2019 10:05 pm

Re: AQR still singing the same song

Post by langlands »

marcopolo wrote: Sat Sep 19, 2020 10:32 pm It is really not any specific strategy that i am questioning. It is the entire notion that really smart people can somehow mine historical data and develop an complex approach that will consistently beat their benchmarks going forward. There is an endless string of individuals and funds that claim to do that, most end up failing. It is especially egregious when they add insult to injury by charging a small fortune for the privilege of investing with them. I also find it interesting that many of the proponents of such complex strategies have convinced themselves that they are so clever, that they feel they need to belittle and question the intelligence of anyone that is not smart enough to grasp just how special their particular complex strategy happens to be.
Yeah, I think this expresses the heart of the disagreement. The problem is that most people think of the "hedge fund" industry as one monolithic entity and it definitely is not. There are a few hedge funds that add a tremendous amount of alpha (market beating returns). The vast majority add minimal alpha (95%+ of them in my estimation). For self-evident economic reasons, the hedge funds that add a tremendous amount of alpha do everything they can to remain secretive and not advertise their funds because they literally have a black box that prints money reliably (> 3.0 Sharpe). The ones that add minimal alpha do everything they can to advertise their funds and attempt to convince investors that they can beat the markets (since they make money from management fees, not market beating trading strategies). Naturally, the big name hedge funds you'll hear about in the news are the non-alpha adding ones since all the forces align for that to happen. Just in case you're skeptical that real alpha exists, Rentech is the prototypical example (they're simply too big and successful to remain hidden).

I think AQR adds minimal alpha. But to their credit, Asness himself is a big proponent behind the "smart beta" movement and does not pretend to have a money printing machine. His whole thing is that he thinks a lot of the "alpha" in the hedge fund industry can actually just be explained by factors. So he's in fact pulling back the curtain so to speak. AQR essentially markets itself as a particularly competent implementer of factor investing and charges a slight premium for this expertise. You can argue whether they actually have this special expertise or not, but you're basically having an argument over 10-20 bps. They aren't promising 15-20% returns per year every year (which a fund like Rentech can do). If you don't think factor investing works, you shouldn't invest with AQR. And I think they're pretty transparent about that. In that respect, they're really pretty honest and up front about what they do. I do find it somewhat amusing that they are held up as a paragon of "quantitative investing" though...
User avatar
Robert T
Posts: 2740
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: AQR still singing the same song

Post by Robert T »

.
If I was to invest in AQR funds it would be along the lines of the following (for the equity portion):

25% = AQR Large Cap Multi-Style [QCELX]
25% = AQR Small Cap Multi-Style [QSMLX]
37% = AQR International Multi-Style [QICLX]
13% = AQR Emerging Multi-Style [QEELX]

Approximate factor loads:
1.00 = Mkt
0.18 = Size
0.36 = Value
0.15 = Momentum

Expense ratio = 0.39

IMO, fairly good in that it gets fairly close to my long-term factor load targets. A bit lower for size and value, but higher for momentum. Don’t know what the quality load would be.
.
User avatar
LadyGeek
Site Admin
Posts: 66405
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: AQR still singing the same song

Post by LadyGeek »

I removed a contentious interchange. As a reminder, see: General Etiquette
This is a moderated forum. We expect this forum to be a place where people can feel comfortable asking questions and where debates and discussions are conducted in civil tones.

Discussions are about issues, not people. If you disagree with an idea, go ahead and marshal all your forces against it. But do not confuse ideas with the person posting them.

At all times we must conduct ourselves in a respectful manner to other posters. Attacks on individuals, insults, name calling, trolling, baiting or other attempts to sow dissension are not acceptable.

If you feel that someone has attacked you or otherwise violated the policies of this forum, do not respond in kind. Instead, please click the report button on the offending post. This is the quickest method to notify the site moderators.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
tarheel
Posts: 409
Joined: Fri Sep 17, 2010 8:33 am

Re: AQR still singing the same song

Post by tarheel »

Robert T wrote: Sun Sep 20, 2020 2:43 am .
If I was to invest in AQR funds it would be along the lines of the following (for the equity portion):

25% = AQR Large Cap Multi-Style [QCELX]
25% = AQR Small Cap Multi-Style [QSMLX]
37% = AQR International Multi-Style [QICLX]
13% = AQR Emerging Multi-Style [QEELX]

Approximate factor loads:
1.00 = Mkt
0.18 = Size
0.36 = Value
0.15 = Momentum

Expense ratio = 0.39

IMO, fairly good in that it gets fairly close to my long-term factor load targets. A bit lower for size and value, but higher for momentum. Don’t know what the quality load would be.
.
Thanks Robert T. This is actually close to how I allocate. Like Ketawa, my entire allocation is in AQR funds apart from 401k restrictions (the rest is with Vanguard).

It blows my mind how people breathlessly badmouth AQR on here. I mean, what's the point? If you think you can better deploy your money elsewhere, fair enough. If you are making a data-driven argument regarding one of their funds, I'm all ears and love to learn, but posting just to insinuate that Cliff is ripping us all off and getting rich is ridiculous.

I can assure you that those of us who allocate heavily to AQR are not idiots. We understand what we are investing in, all the way to (dare I mention!) QSPIX or QMHIX. We are paying what we think is a fair fee for what we think will be worth it in the long run. We used to post a lot more. I for one have a young family and my free time isn't what it used to be. I am right on track for a comfortable early retirement with my AQR-centric portfolio.
User avatar
tarheel
Posts: 409
Joined: Fri Sep 17, 2010 8:33 am

Re: AQR still singing the same song

Post by tarheel »

Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
Awesome to hear Rick - and love the podcast by the way, thanks so much for doing it. Excellent stuff.
User avatar
nedsaid
Posts: 13849
Joined: Fri Nov 23, 2012 12:33 pm

Re: AQR still singing the same song

Post by nedsaid »

langlands wrote: Sun Sep 20, 2020 12:21 am
marcopolo wrote: Sat Sep 19, 2020 10:32 pm It is really not any specific strategy that i am questioning. It is the entire notion that really smart people can somehow mine historical data and develop an complex approach that will consistently beat their benchmarks going forward. There is an endless string of individuals and funds that claim to do that, most end up failing. It is especially egregious when they add insult to injury by charging a small fortune for the privilege of investing with them. I also find it interesting that many of the proponents of such complex strategies have convinced themselves that they are so clever, that they feel they need to belittle and question the intelligence of anyone that is not smart enough to grasp just how special their particular complex strategy happens to be.
Yeah, I think this expresses the heart of the disagreement. The problem is that most people think of the "hedge fund" industry as one monolithic entity and it definitely is not. There are a few hedge funds that add a tremendous amount of alpha (market beating returns). The vast majority add minimal alpha (95%+ of them in my estimation). For self-evident economic reasons, the hedge funds that add a tremendous amount of alpha do everything they can to remain secretive and not advertise their funds because they literally have a black box that prints money reliably (> 3.0 Sharpe). The ones that add minimal alpha do everything they can to advertise their funds and attempt to convince investors that they can beat the markets (since they make money from management fees, not market beating trading strategies). Naturally, the big name hedge funds you'll hear about in the news are the non-alpha adding ones since all the forces align for that to happen. Just in case you're skeptical that real alpha exists, Rentech is the prototypical example (they're simply too big and successful to remain hidden).

I think AQR adds minimal alpha. But to their credit, Asness himself is a big proponent behind the "smart beta" movement and does not pretend to have a money printing machine. His whole thing is that he thinks a lot of the "alpha" in the hedge fund industry can actually just be explained by factors. So he's in fact pulling back the curtain so to speak. AQR essentially markets itself as a particularly competent implementer of factor investing and charges a slight premium for this expertise. You can argue whether they actually have this special expertise or not, but you're basically having an argument over 10-20 bps. They aren't promising 15-20% returns per year every year (which a fund like Rentech can do). If you don't think factor investing works, you shouldn't invest with AQR. And I think they're pretty transparent about that. In that respect, they're really pretty honest and up front about what they do. I do find it somewhat amusing that they are held up as a paragon of "quantitative investing" though...
A few things here. First, you can learn an awful lot from people that you disagree with. Some folks have never learned that and seem to have devoted themselves to running such folks off. Such a shame as this limits discussion and the sheer joy of reasoned debate. Second, not everybody with a contrary opinion is out to rip you off or to deceive you. One such person made controversial recommendations but put substantial amounts of their own money into them. The spirited disagreements were fun until they started getting personal. Again, if you keep your eyes and ears open, one might learn something. I have no money at AQR but I will be delighted to hear what Cliff Asness has to say when Rick Ferri interviews him. I will be all ears but that doesn't mean that I will rush to put all my money in AQR afterwards. I want to hear what a very intelligent market participant has to say about investing and the markets.
A fool and his money are good for business.
User avatar
nedsaid
Posts: 13849
Joined: Fri Nov 23, 2012 12:33 pm

Re: AQR still singing the same song

Post by nedsaid »

tarheel wrote: Sun Sep 20, 2020 8:04 am
Rick Ferri wrote: Fri Sep 18, 2020 7:25 pm Cliff Asness agreed to be my November guest on the Bogleheads on Investing podcast. Bogleheads will be able to ask him intelligent questions that would benefit the community. I will post more next month.

Rick Ferri
Awesome to hear Rick - and love the podcast by the way, thanks so much for doing it. Excellent stuff.
Thank you Rick, I love your podcasts too and I thank you for convincing Cliff Asness to be your guest.
A fool and his money are good for business.
Post Reply