QSPIX - thoughts on interesting fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 19, 2015 5:32 pm

Well, this is certainly not what you want if targeting equity factors. You can presumably have higher loadings (for the risk taken) at a lower price with the equity market neutral fund (QMNIX), which isn't diluting the money pool by spending most of it chasing factors in other asset classes.

That said, AQR would say that there is a relationship and correlation (though seemingly not incredibly high) between the same factor across asset classes, so in some roundabout way you can think of the rest as targeting equity factors even without using equities?

User avatar
nisiprius
Advisory Board
Posts: 34135
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: QSPIX - thoughts on interesting fund

Post by nisiprius » Fri Jun 19, 2015 8:38 pm

To the people who only plan to use a small percentage allocation, such as Larry Swedroe's 3%, what, really, do you expect from this?

Let's suppose that we had a fund that was guaranteed to have, going forward, almost exactly the same return and standard deviation as Total Stock, and 0% correlation. We would, then, see something like this.

Image

If we had about a 50% mix of the two, the 0% correlation would cause a large reduction in risk. If we decide that we want our portfolio to have 10% standard deviation, the effect of replacing stocks with a 50/50 mix of stocks and the second asset is to create a 1.55% increase in annual return. That's pretty interesting.

But we are talking about something like 3% allocation. That means we are really talking about the dashed blue line. I am too lazy to do the precise calculations, but reading the chart by eyeball and doing some approximate calculations, it looks as if we get an 0.20% increase in annual return from a 3% allocation.

Of course with a 6% allocation we would get perhaps 0.40%, and if we were happy with about the same volatility as 100% stocks the gain would be 0.80%.

Again, that is from a perfectly behaved "second asset" that has just as high a return as stocks, with a reliable 0.00 correlation.
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.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 19, 2015 9:21 pm

Personally I kind of agree with small allocations being mostly noise and have been suggesting say 10% as a starting point (quite possibly lower, maybe higher? but not a whole lot higher). I know Larry Swedroe said something about 3%, but his equity allocation is pretty low, if I understand it correctly. Still, 3% of anything doesn't do much good or bad.

For what it's worth, the suggested 7% annual returns (was this over T-bills or not? I guess it's the same right now) and 10% standard deviation with practically 0 correlation to stocks would make it better than the stock proxy in the above example. That's kind of the best-case scenario, though.

To be honest, I think a close-to-0 correlation with stocks isn't all that farfetched. Correlation may creep up a bit under the right kind of stress (currency carry, for example, tends to do bad in flights to safety that hit stocks too, though maybe loading on defensive stocks balances that), but over the backtest and since inception it's been meaningfully about 0. That's about what you should expect from the net hedging most of the time.

It's the expected return that nobody can really figure out. If the future looks similar to the past, it could well be over 7% or whatever that people are quoting. If this stuff stops working somehow, it could well be significantly negative after costs and fees. I'd wager on something in the middle but without great confidence.

User avatar
tarheel
Posts: 307
Joined: Fri Sep 17, 2010 8:33 am

Re: QSPIX - thoughts on interesting fund

Post by tarheel » Sat Jun 20, 2015 6:06 am

nisiprius wrote: Of course with a 6% allocation we would get perhaps 0.40%, and if we were happy with about the same volatility as 100% stocks the gain would be 0.80%.


I have 6.25% QSPIX in a 78.75% stocks/15% bond portfolio. I have a 30 year horizon to retirement. Your numbers sound great to me. :)

User avatar
nisiprius
Advisory Board
Posts: 34135
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: QSPIX - thoughts on interesting fund

Post by nisiprius » Sat Jun 20, 2015 8:16 am

tarheel wrote:
nisiprius wrote: Of course with a 6% allocation we would get perhaps 0.40%, and if we were happy with about the same volatility as 100% stocks the gain would be 0.80%.
I have 6.25% QSPIX in a 78.75% stocks/15% bond portfolio. I have a 30 year horizon to retirement. Your numbers sound great to me. :)
Sure, guarantee me that the return, standard deviation, and low correlation will persist and I'll buy a chunk of it tomorrow. The devil of it is all the things that we are told "have" certain characteristics ("have" is apparently a technical term in investing that means "has had"), that mysteriously lose those characteristics just about the time I start seeing articles about them.
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
Robert T
Posts: 2446
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T » Sat Jun 20, 2015 9:29 am

.
The results of backtested portfolios depend on the returns/SD and correlations used for the style premia series. Using 10% annualized return/10 standard deviation as per the AQR paper, with zero correlation with stocks and bond yielded (small) mean-variance efficiency improvements for an already global small cap and value tilted portfolio over 1990-2013. Using the 7% annualized return expectation (net of fees) [instead of 10%] from Larry's earlier article did not improve mean-variance efficiency for an already tilted portfolio [at least in my simulations].

I also get similar results to those in the paper on Understaning Style Premia where both returns are higher and the SD is lower relative to a 60/40 total stock/bond portfolio if a 30% allocation of the style premium series is added, and using a 10% return/10 SD for the series. If the 7% return is used, the benefit is much smaller (lower return and SD, but still higher Sharpe ratio). A long-only titled portfolio provided better mean-variance efficiency than adding the style premium a series to a market portfolio.

A key reason for the above results is that 5 yr-T notes performed well over this period with an annualized return of 5.9% and standard deviation of 6, with a -0.39 annual return correlation with the (globally diversified small cap and value tilted) stock portfolio. i.e. 5 yr-T notes provided more effective diversification over this period. Going forward the returns of 5 yr T-notes may be lower with smaller diversification benefit. There seems to be a wider variation in 'estimated' returns of the style premia series than with long-only series. For example the raw data shows a return of 25% while the 'adjusted' data shows a return of 10%. Thats as large adjustment and with it likely large 'standard errors' or uncertainty of actual long-term returns. Time will tell.

The greatest diversification benefit of the style premia series seems to be in inflationary recessions (as per earlier point) - this was about 8% of the time from 1926. Anyone's guess if this will be a larger share in the future than the past.

Obviously no guarantees.

Robert
.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Sat Jun 20, 2015 11:46 am

For that matter, can anybody explain the difference in the numbers between "Investing with Style" and "Understanding Style Premia"? Both cover 1990-2013 and give composite and individual style performance numbers, but they're significantly different. "Investing with Style" puts the composite at 17.4% over the risk-free rate and "Understanding Style Premia" at 25.9%. The latter is explicitly before fees and transaction costs, though that seems to also be the case for the former?

http://scholar.google.com/scholar_url?u ... s=1920x911
https://www.aqr.com/~/media/files/paper ... -style.pdf

Also, the 7% from Larry's article and elsewhere.

Now, if results are more like 3-4% that is still worth including at least in the short term with most expectations for stocks and certainly of bonds. In a scenario where say bonds return 2% with 5% standard deviation, a different asset with low correlation and 3% return with 10% standard deviation is useful in some smaller role alongside that.

To me, if we're going back and plugging in style premia results back many years into a simulated backtest, it's more fair to use the higher estimates closer to the raw historical results. In any case, if you're accounting for transaction costs (a good idea), you need to go back and do that for all the other asset classes, though how you treat transaction costs of small international companies decades ago is obviously very subjective. More usefully, if projecting in the future, you definitely need to use lower bond returns given current yields, and if you want to discount one fund's "it probably won't work as well in the future" (the adjusted numbers like the 10% and 7% have this baked in) then you should probably do the same for say small and value factors.

countmein
Posts: 404
Joined: Fri Dec 06, 2013 9:10 pm

Re: QSPIX - thoughts on interesting fund

Post by countmein » Sat Jun 20, 2015 3:04 pm

nisiprius wrote:To the people who only plan to use a small percentage allocation, such as Larry Swedroe's 3%, what, really, do you expect from this?


good question. I expect it to beat the pants off my stock investments, at least for the next decade, so any amount is good to have.

User avatar
wbond
Posts: 1070
Joined: Wed Dec 10, 2008 6:55 pm

Re: QSPIX - thoughts on interesting fund

Post by wbond » Sat Jun 20, 2015 3:14 pm

Robert T wrote:.
The results of backtested portfolios depend on the returns/SD and correlations used for the style premia series. Using 10% annualized return/10 standard deviation as per the AQR paper, with zero correlation with stocks and bond yielded (small) mean-variance efficiency improvements for an already global small cap and value tilted portfolio over 1990-2013. Using the 7% annualized return expectation (net of fees) [instead of 10%] from Larry's earlier article did not improve mean-variance efficiency for an already tilted portfolio [at least in my simulations].

I also get similar results to those in the paper on Understaning Style Premia where both returns are higher and the SD is lower relative to a 60/40 total stock/bond portfolio if a 30% allocation of the style premium series is added, and using a 10% return/10 SD for the series. If the 7% return is used, the benefit is much smaller (lower return and SD, but still higher Sharpe ratio). A long-only titled portfolio provided better mean-variance efficiency than adding the style premium a series to a market portfolio.

A key reason for the above results is that 5 yr-T notes performed well over this period with an annualized return of 5.9% and standard deviation of 6, with a -0.39 annual return correlation with the (globally diversified small cap and value tilted) stock portfolio. i.e. 5 yr-T notes provided more effective diversification over this period. Going forward the returns of 5 yr T-notes may be lower with smaller diversification benefit. There seems to be a wider variation in 'estimated' returns of the style premia series than with long-only series. For example the raw data shows a return of 25% while the 'adjusted' data shows a return of 10%. Thats as large adjustment and with it likely large 'standard errors' or uncertainty of actual long-term returns. Time will tell.

The greatest diversification benefit of the style premia series seems to be in inflationary recessions (as per earlier point) - this was about 8% of the time from 1926. Anyone's guess if this will be a larger share in the future than the past.

Obviously no guarantees.

Robert
.


Robert, all of your posts are sincerely a joy to read. I quote this one, only since it is the most current, but the prior several on this topic are all cogently and clearly delivered.

I read the Ilmanen book when it was first published (twice, as is my rule with anything penned by a Finn) and enjoyed it more from the standpoint of theory than practice. That said, I do share a curiosity with this fund, and I think you do well in analyzing the trade-offs.

One additional concern that I didn't see skimming the thread (but, c'mon, eight pages!) is a "behavioral" concern. With other strategic decisions that I personally make in constructing my asset allocation (intl, em, s/v, etc), whether I stick with them and buy low when they "fail" is up to me and my "behavior." But with QSPIX, what if I am perfectly content to rebalance into the fund when times are tough for its approach but the "market" sells out and AQR ditches the current strategy?

All backtesting of strategies assumes implicitly good investing "behavior," which we know in real life is a virtue only possessed by a minority. Yet the majority have a significant impact on the survival of any single fund such as this one.

User avatar
nisiprius
Advisory Board
Posts: 34135
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: QSPIX - thoughts on interesting fund

Post by nisiprius » Sun Jun 21, 2015 7:21 am

The general question, with all strategies that a) are validated only by very long runs of backtested data, and b) show periods of of outperformance and underperformance that have, in the past, lasted for over a decade, is: do I personally have the depth of conviction (and life expectancy) needed to stay the course for long enough for the long-term statistics to prevail--assuming that they do prevail, and that I am not simply the victim of sampling error or an actual change in the markets (driven by changes in laws, or government decisions, perhaps).

The temptation to think that you can "try and see" is an invitation to jumping onto fads at the bad times and then bailing at bad times. On the basis of their supporting past data, most investing strategies are not worth pursuing unless you are prepared to make something resembling a lifetime commitment.

Even with 3%, 5%, or 10%, you need to think "this IS my fun money" and not pursue any other "fun money" investments... and even with a small allocation, when you go in, you should make some judgement as to the length of time you need to give the strategy in order for it to have a decent chance of success--and then hang in for that long. In other words, "this is my fun money... for the next two decades... and I will ignore the next article saying the next interesting idea."

That's not specific to QSPIX, and actually I haven't looked at the backtested data so I don't actually know whether or not it has had dishearteningly long periods of underperformance. And of course there's no difference between QSPIX and Total Stock with regard to the commitment aspect, except that the farther you are out of the mainstream, the deeper the conviction needed to stay the course.
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
wbond
Posts: 1070
Joined: Wed Dec 10, 2008 6:55 pm

Re: QSPIX - thoughts on interesting fund

Post by wbond » Sun Jun 21, 2015 10:33 am

nisiprius wrote:The general question, with all strategies that a) are validated only by very long runs of backtested data, and b) show periods of of outperformance and underperformance that have, in the past, lasted for over a decade, is: do I personally have the depth of conviction (and life expectancy) needed to stay the course for long enough for the long-term statistics to prevail--assuming that they do prevail, and that I am not simply the victim of sampling error or an actual change in the markets (driven by changes in laws, or government decisions, perhaps).

The temptation to think that you can "try and see" is an invitation to jumping onto fads at the bad times and then bailing at bad times. On the basis of their supporting past data, most investing strategies are not worth pursuing unless you are prepared to make something resembling a lifetime commitment.

Even with 3%, 5%, or 10%, you need to think "this IS my fun money" and not pursue any other "fun money" investments... and even with a small allocation, when you go in, you should make some judgement as to the length of time you need to give the strategy in order for it to have a decent chance of success--and then hang in for that long. In other words, "this is my fun money... for the next two decades... and I will ignore the next article saying the next interesting idea."

That's not specific to QSPIX, and actually I haven't looked at the backtested data so I don't actually know whether or not it has had dishearteningly long periods of underperformance. And of course there's no difference between QSPIX and Total Stock with regard to the commitment aspect, except that the farther you are out of the mainstream, the deeper the conviction needed to stay the course.


Correct, but it's even worse here. You may have the conviction in bad times, but the fund folds or changes course leaving you with nothing to replace it with. One's behavior is bad enough, here you are also indirectly tied to market behavior.

User avatar
Robert T
Posts: 2446
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T » Sun Jun 21, 2015 11:10 am

wbond wrote:One additional concern that I didn't see skimming the thread (but, c'mon, eight pages!) is a "behavioral" concern. With other strategic decisions that I personally make in constructing my asset allocation (intl, em, s/v, etc), whether I stick with them and buy low when they "fail" is up to me and my "behavior." But with QSPIX, what if I am perfectly content to rebalance into the fund when times are tough for its approach but the "market" sells out and AQR ditches the current strategy?

All backtesting of strategies assumes implicitly good investing "behavior," which we know in real life is a virtue only possessed by a minority. Yet the majority have a significant impact on the survival of any single fund such as this one.

Yes, a risk. I am not sure how large or what impact possible large redemptions (and associated unwinding of positions) in fund like QSPIX would have on portfolio performance (not factored into simulated returns). From Illmanen's book:

Admittedly, leveraged positions come with a greater likelihood of having to de-lever in bad times at fire sale prices. More generally, if you monetize diversification gains via leverage, and diversification fails in bad times—recall the correlation spikes in 1998 and 2008—a broadly diversified portfolio can experience sharper losses than a simple equity-concentrated portfolio.

Agree that it is an excellent book.

Robert
.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Sun Jun 21, 2015 11:15 am

nisiprius wrote:The general question, with all strategies that a) are validated only by very long runs of backtested data, and b) show periods of of outperformance and underperformance that have, in the past, lasted for over a decade, is: do I personally have the depth of conviction (and life expectancy) needed to stay the course for long enough for the long-term statistics to prevail--assuming that they do prevail, and that I am not simply the victim of sampling error or an actual change in the markets (driven by changes in laws, or government decisions, perhaps).

The temptation to think that you can "try and see" is an invitation to jumping onto fads at the bad times and then bailing at bad times. On the basis of their supporting past data, most investing strategies are not worth pursuing unless you are prepared to make something resembling a lifetime commitment.

Even with 3%, 5%, or 10%, you need to think "this IS my fun money" and not pursue any other "fun money" investments... and even with a small allocation, when you go in, you should make some judgement as to the length of time you need to give the strategy in order for it to have a decent chance of success--and then hang in for that long. In other words, "this is my fun money... for the next two decades... and I will ignore the next article saying the next interesting idea."

That's not specific to QSPIX, and actually I haven't looked at the backtested data so I don't actually know whether or not it has had dishearteningly long periods of underperformance. And of course there's no difference between QSPIX and Total Stock with regard to the commitment aspect, except that the farther you are out of the mainstream, the deeper the conviction needed to stay the course.

Here, the backtest for the strategy as a whole really only starts from 1990, but for many style/asset class combinations it goes back a lot further than that.

In any case, here's AQR's own simulated numbers from "Investing with Style" for each of the four styles (remember, each is implemented across multiple asset classes), which are combined and rebalanced in the real thing:
Image
Obviously those four styles and not something else were chosen because those are historically known to work more often than not. It's a "validation" of ideas using the data that generated the ideas (to some extent; you could well call value investing for example much older than 1990) so of course it looks good there. For those who remember currency carry getting blown up in the financial crisis, carry in other asset classes as defined softened the blow.

Part of the appeal of going to multiple asset classes and multiple styles is that all of them won't be losing money simultaneously. Usually. In any case, value and momentum are pretty persistently negatively correlated.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Mon Jun 22, 2015 9:51 am

nisiprius wrote:The general question, with all strategies that a) are validated only by very long runs of backtested data, and b) show periods of of outperformance and underperformance that have, in the past, lasted for over a decade, is: do I personally have the depth of conviction (and life expectancy) needed to stay the course for long enough for the long-term statistics to prevail--assuming that they do prevail, and that I am not simply the victim of sampling error or an actual change in the markets (driven by changes in laws, or government decisions, perhaps).

The temptation to think that you can "try and see" is an invitation to jumping onto fads at the bad times and then bailing at bad times. On the basis of their supporting past data, most investing strategies are not worth pursuing unless you are prepared to make something resembling a lifetime commitment.

Even with 3%, 5%, or 10%, you need to think "this IS my fun money" and not pursue any other "fun money" investments... and even with a small allocation, when you go in, you should make some judgement as to the length of time you need to give the strategy in order for it to have a decent chance of success--and then hang in for that long. In other words, "this is my fun money... for the next two decades... and I will ignore the next article saying the next interesting idea."

That's not specific to QSPIX, and actually I haven't looked at the backtested data so I don't actually know whether or not it has had dishearteningly long periods of underperformance. And of course there's no difference between QSPIX and Total Stock with regard to the commitment aspect, except that the farther you are out of the mainstream, the deeper the conviction needed to stay the course.


I confess, if this fund delivers negative real returns after its first 5 years I will have a hard time hanging on to it and I'm not even sure I should hang on to it if that is the case. Is it poor implementation and cost drag/fees or not? I have much more conviction in my other funds and this is not an issue with them. Either way, a 10% holding will not make or break an investor, but I hope it's positive.

In defense of new fads, give some credit here. I do not think seasoned Bogleheads are quick to jump on alternative investments. I was a huge skeptic myself and never envisioned holding an allocation to alternatives. This fund is unique though in strategy and I researched it for a while before taking the plunge. The decision for me to buy QSPIX was not taken willy nilly.
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Wed Jun 24, 2015 7:18 am

matjen wrote:
It is also entertaining. What would I do with my time if I didn't have this thread? :happy


Looks like this thread has run its course. What happened to stay the course? Back to the shadows I go until some more QSPIX stuff comes out, or a new factor is invented, or a good yoga thread comes along....
There are no guarantees, only probabilities.

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

Re: QSPIX - thoughts on interesting fund

Post by matjen » Thu Jun 25, 2015 2:10 pm

grap0013 wrote:
matjen wrote:
It is also entertaining. What would I do with my time if I didn't have this thread? :happy


Looks like this thread has run its course. What happened to stay the course? Back to the shadows I go until some more QSPIX stuff comes out, or a new factor is invented, or a good yoga thread comes along....



Don't worry Grap...I started another one! :twisted:
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
Yesterdaysnews
Posts: 424
Joined: Sun Sep 14, 2014 1:25 pm

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Mon Jun 29, 2015 6:34 pm

Down 0.61% today when S&P 500 down 2.09% - not bad.

Random Walker
Posts: 2051
Joined: Fri Feb 23, 2007 8:21 pm

Re: QSPIX - thoughts on interesting fund

Post by Random Walker » Mon Jun 29, 2015 10:26 pm

Ya, one day is meaningless but interesting nonetheless. The major indices down 2%, Foreign indices down about 2.5%, my muni bonds up 0.2%, CCFs down 0.3%, QSPIX down 0.6%. Based on this 1 day sample, CCFs and QSPIX are good diversifiers. The best diversifier, especially for the cost is bonds.

Dave

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

Re: QSPIX - thoughts on interesting fund

Post by HomerJ » Mon Jun 29, 2015 10:40 pm

packer16 wrote:The fees are another big issue for me. I am reluctant to pay 1% for a known strategy with a skilled implementer (Sequoia) so why should I pay 1.5% to strategy that is not totally known and the implementer who has a short track record with this strategy and had a blow up in one of his funds (AQR Absolute Return) in the last crisis


I really don't understand why people are so willing to pay high expenses for an unproven strategy... A strategy which is not even disclosed.

"Invest with us! We have 20 strategies... At least a few of them should work! No, we can't tell you what they are!"
Last edited by HomerJ on Mon Jun 29, 2015 11:11 pm, edited 1 time in total.

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

Re: QSPIX - thoughts on interesting fund

Post by HomerJ » Mon Jun 29, 2015 11:09 pm

Yesterdaysnews wrote:Down 0.61% today when S&P 500 down 2.09% - not bad.


Total Bond Market Index Fund was UP 0.66% today...

I'm really not sure what you people think you are accomplishing here.

Plain old stocks/bonds seems to be working just fine for diversification today.

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

Re: QSPIX - thoughts on interesting fund

Post by HomerJ » Mon Jun 29, 2015 11:46 pm

packer16 wrote:http://www.wsj.com/articles/SB124303417991748685.


Did you guys read this?

For Mr. Asness, the losses were especially painful. He and several other partners started the firm in 1998 after leaving Goldman Sachs Group Inc., where they had launched the Global Alpha fund. He built his returns on the idea that markets for stocks, bonds, currencies and other securities could be irrational for a period of time but always came back to normal. By making lots of small bets, it could produce profits no matter which way the markets moved.

But last year's turmoil hurt AQR's complex trading strategy. As rumors swirled about the size of AQR's losses on a message board on the Wall Street blog Dealbreaker on Dec. 4, Mr. Asness, sitting in his office at AQR's headquarters in Greenwich, Conn., responded.

"This is Cliff Asness," he began, launching into a diatribe against the "cowardly posts" and "little liars" spreading rumors. He posted the message on the message board, and quickly realized he had made a mistake.

"It was stupid," Mr. Asness said. While he said the rant was an extreme reaction, the hedge-fund manager is known for speaking his mind.

Mr. Asness, 42 years old, has seen a number of changes over his long career managing money. But he had never seen a market quite like the one that pounded his hedge fund in the past year and a half.


So, his fund in 2008 failed because something new happened.

I hope everything in the future is exactly like the past, because if it's something Cliff Asness has never seen before, you might be in for a rough ride.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Tue Jun 30, 2015 12:02 am

HomerJ wrote:
packer16 wrote:The fees are another big issue for me. I am reluctant to pay 1% for a known strategy with a skilled implementer (Sequoia) so why should I pay 1.5% to strategy that is not totally known and the implementer who has a short track record with this strategy and had a blow up in one of his funds (AQR Absolute Return) in the last crisis


I really don't understand why people are so willing to pay high expenses for an unproven strategy... A strategy which is not even disclosed.

"Invest with us! We have 20 strategies... At least a few of them should work! No, we can't tell you what they are!"

It's obviously for the uniqueness of coverage in one fund. I can understand and agree with a high degree of skepticism around multiple issues here, but what people are looking or hoping for shouldn't be any mystery. They're looking for different (multiple) sources of returns with low correlation to other assets that provide returns.

Investing is about playing the odds with your information and convictions. We all think stocks are likely to go up in the long run or we wouldn't invest in them, even if we all don't explicitly assign estimated likelihoods to different scenarios. With the example of Sequoia, while one investor sees an attractive strategy (albeit with unknown details about the exact gut feelings and analyses that go into individual security picks, trading regimen, and so on) with a history of success and perhaps some likelihood of that continuing in the future, another who more heavily weighs historical information about the lack of US stock mutual fund performance persistence will be skeptical and stay away despite the fund's track record. There are always unknowns with a fund's operation to some degree, never mind questions about how the markets work.

With the style premia fund there are not judgment calls about individual markets, securities, and directions. In many meaningful ways, there are a lot more strategy disclosures here than for most active funds. There is enough information, assuming they're not lying, to know something substantial about the driving principles and formulations at a high level. Regardless, people shouldn't invest based only on what is known, but weighing all the different aspects, both known and unknown. You have to guess whether or not the returns will be positive or not, by how much, and if they'll actually have low correlation to stocks, bonds, and the like.

HomerJ wrote:So, his fund in 2008 failed because something new happened.

I hope everything in the future is exactly like the past, because if it's something Cliff Asness has never seen before, you might be in for a rough ride.

I've said it all along that everyone should expect models to not fit reality well sometimes, and for quant funds to take some unexpected lumps here and there on top of all the ones implied by the targeted volatility and the models. Whether or not the pros outweigh the cons is another matter, one for people to decide. And again, any given alternative does not have to be better than stocks or bonds to improve a portfolio of stocks and bonds, because of the nature of diversification. (It just can't be a lot worse.)

In any case, it's been brought up many times that the amount of risk taken in Absolute Return was greater (exacerbating the magnitude of the drawdown, making it not comparable), and with more strategies here there should be less of a chance for as many things to go sour simultaneously.


Random Walker wrote:Ya, one day is meaningless but interesting nonetheless. The major indices down 2%, Foreign indices down about 2.5%, my muni bonds up 0.2%, CCFs down 0.3%, QSPIX down 0.6%. Based on this 1 day sample, CCFs and QSPIX are good diversifiers. The best diversifier, especially for the cost is bonds.

Dave

Just pulling up a few more dates, based on Morningstar data for days in 2014 (2/3, 4/10, 9/25, 10/9, 10/13) with US stock returns of -2.43%, -2.20%, -1.58%, -2.12%, and -1.58%, the AQR style premia fund returned -0.20%, -0.39%, -0.29%, -0.10%, and -0.29%. I checked a few other bad stock days (worse than 1%) and the AQR fund was down a bit for those too.

This is with practically 0 correlation to stocks over the period, while attempting to be market neutral, and having no market beta for the period based on a regression. Looks like something(s) in there might be sensitive to market freakouts to some limited degree, even if longer-term behavior in the backtest and in live behavior hasn't yet shown a real relationship to stocks.

One naive expectation for a market neutral, hedged fund would be independence from stock returns (which as an aside would also follow from zero correlation and normal distributions), so performance conditioned on stocks having a good or bad day wouldn't be different from the unconditional performance. Note that nobody should actually expect this even with zero market exposure because the other stuff could well be systemically correlated to stocks. In any case, this very naive expectation seems wrong in practice, seeing as bad stock days seem to make drawdowns here more likely. I wonder if this is just from carry, or what's causing it.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Tue Jun 30, 2015 7:36 am

HomerJ wrote:
Yesterdaysnews wrote:Down 0.61% today when S&P 500 down 2.09% - not bad.


Total Bond Market Index Fund was UP 0.66% today...

I'm really not sure what you people think you are accomplishing here.

Plain old stocks/bonds seems to be working just fine for diversification today.


Treasury bonds were up even a little bit more. They are still your best diversifier to stocks, no doubt. Negative correlation. QSPIX has no correlation to stocks. Not even to butter production in Bangladesh. Sometimes QSPIX goes up when stocks are down and sometimes it goes down when stocks are down. There are many days when stocks are either down or flat, treasuries are down or flat, and QSPIX is up. If it can happen on a daily basis it can happen on an annual basis as well. That's diversification whether you like to admit it or not.
There are no guarantees, only probabilities.

randomguy
Posts: 4884
Joined: Wed Sep 17, 2014 9:00 am

Re: QSPIX - thoughts on interesting fund

Post by randomguy » Tue Jun 30, 2015 8:43 am

grap0013 wrote:
HomerJ wrote:
Yesterdaysnews wrote:Down 0.61% today when S&P 500 down 2.09% - not bad.


Total Bond Market Index Fund was UP 0.66% today...

I'm really not sure what you people think you are accomplishing here.

Plain old stocks/bonds seems to be working just fine for diversification today.


Treasury bonds were up even a little bit more. They are still your best diversifier to stocks, no doubt. Negative correlation. QSPIX has no correlation to stocks. Not even to butter production in Bangladesh. Sometimes QSPIX goes up when stocks are down and sometimes it goes down when stocks are down. There are many days when stocks are either down or flat, treasuries are down or flat, and QSPIX is up. If it can happen on a daily basis it can happen on an annual basis as well. That's diversification whether you like to admit it or not.


Treasury bonds don't have negative correlation with stocks. Look at the bond return of the last 5 years. If they were negatively correlated, bonds would have huge losses. They don't. The only thing I am aware of that is negatively correlated with stocks is shorting the stock market:)

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Tue Jun 30, 2015 10:15 am

randomguy wrote:
Treasury bonds don't have negative correlation with stocks. Look at the bond return of the last 5 years. If they were negatively correlated, bonds would have huge losses. They don't. The only thing I am aware of that is negatively correlated with stocks is shorting the stock market:)


Au contraire mon frère, they are definitely negatively correlated: https://www.portfoliovisualizer.com/ass ... S%2C+VFITX

Correlations are directional, but also remember, there is a magnitude as well as a timing component to correlations as well. Treasuries go extremely negatively correlated to stocks just when you need them to like in 2008. They typically go up when stocks have a big down day as well. Timing, magnitude, and direction all have a role here when talking about correlations. This is a hard topic to explain simplistically.
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Tue Jun 30, 2015 10:19 am

One magnitude of correlation example:

Let's say US stocks and EMs have a positive correlation of 0.68 and you get yearly returns of:

2012
US 10%
EM 1%

2013
US -2%
EM -18%

2014
US 5%
EM 3%

Correlations were all in the same direction (positive vs. negative returns) but US returns are in the green while EM returns are in the red in this example. I hope this helps others conceptualize this topic.
There are no guarantees, only probabilities.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Tue Jun 30, 2015 10:28 am

grap0013 wrote:
randomguy wrote:
Treasury bonds don't have negative correlation with stocks. Look at the bond return of the last 5 years. If they were negatively correlated, bonds would have huge losses. They don't. The only thing I am aware of that is negatively correlated with stocks is shorting the stock market:)


Au contraire mon frère, they are definitely negatively correlated: https://www.portfoliovisualizer.com/ass ... S%2C+VFITX

Correlations are directional, but also remember, there is a magnitude as well as a timing component to correlations as well. Treasuries go extremely negatively correlated to stocks just when you need them to like in 2008. They typically go up when stocks have a big down day as well. Timing, magnitude, and direction all have a role here when talking about correlations. This is a hard topic to explain simplistically.

The thing to remember is that the numerator for correlation uses (X - E(X)) terms, not raw X. In other words, it's the deviation relative to the mean that is considered, not the direction.

Johno
Posts: 1883
Joined: Sat May 24, 2014 4:14 pm

Re: QSPIX - thoughts on interesting fund

Post by Johno » Tue Jun 30, 2015 10:37 am

Over the long term treasury bond prices have had close to zero correlation to stocks not negative (though as also explained, the correlation could be negative without meaning treasuries generally have to lose money if stocks generally go up, and has been negative in recent years). Assumption of negative correlation even in bad periods for the stock market is somewhat of a recency bias. It hasn't been as true when the stock market sells off on discomfort with high inflation or high real short rates from central bank, and the correlation could turn very positive in a broad crisis of confidence in developed country debt (which doesn't exist in modern historical data but that doesn't make it impossible).

The idea of QSPIX is to 'harvest risk premia' not correlated with the equity risk premium. In the pure CAPM world the 'riskless asset' is something like a T-bill with practically zero duration as well as presumed zero credit risk, which is also nearly 100% realistic for a very short term instrument only assuming there's anywhere else you could put the money, at least you know you can take it out of T-bills (say) without market loss if you only perceive a relatively slow moving deterioration in the govt's credit when everyone else does. But in practical/popularized CAPM ('you only need bonds and stocks') the investor definitely takes duration risk in buying longer term bonds, to get any return; and also if govt credit deteriorated in that case you couldn't expect to get out of long term govt bonds unscathed unless you detected the deterioration before everybody else. But, the duration risk part at least is basically uncorrelated to the equity risk premium. The credit risk part is hard to analyze since history says it doesn't exist at all.

Anyway in practice even considering govt bonds of any term truly credit 'riskless' the actual return on today's curve is coming from taking duration risk, harvesting the term premium. If you refused to take that risk you'd now get practically no return in US govt instruments per se, though you could get a 1% nominal, minus some fraction of 1% real (at recent inflation) pre tax return with no duration risk and nearly pure govt credit risk in best paying FDIC insured bank accounts. IOW the concept of QSPIX really only says there might be alternative sources of return (basically) uncorrelated to the equity risk premium than the one you're already taking to make anything in govt bonds, the term premium. And as repeated now ad nauseam the argument in favor of such a fund is *not* that it would entirely replace either stocks or govt bonds, but rather just diversify your sources of risk taking.

But again, will these other sources of uncorrelated return be as durable in the future as the term premium, or equity risk premium itself? (though we don't actually have a future gtee on the size of either of those two premia either). And you do have to pay significant cost, and have some black box issues (albeit often overstated on these threads, sometimes ridiculously so) to access the other putative sources of return via QSPIX (or any other way, AFAIK).

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Thu Jul 02, 2015 7:50 am

1 week returns:

My entire portfolio (all equities): -1.55%
intermed term treasuries VFITX: -0.15%
QSPIX: +0.82%

What is not to like?! :D
There are no guarantees, only probabilities.

User avatar
Yesterdaysnews
Posts: 424
Joined: Sun Sep 14, 2014 1:25 pm

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Wed Jul 08, 2015 5:23 pm

Performed poorly today:

Down 1.62% vs. 1.67% for S&P index.

User avatar
nisiprius
Advisory Board
Posts: 34135
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: QSPIX - thoughts on interesting fund

Post by nisiprius » Wed Jul 08, 2015 5:36 pm

grap0013 wrote:Au contraire mon frère, they are definitely negatively correlated: https://www.portfoliovisualizer.com/ass ... S%2C+VFITX ...
No, they are not.

That's for a short, recent period of time. "Asset correlations for time period 01/28/2011 - 06/29/2015 based on daily returns."

For 1926-2009, inflation-adjusted, the correlation of large-company stocks is +0.23 with long-term government bonds, +0.11 with long-term government bonds, +0.08 for intermediate-term government bond. Nominal, +0.17 with long-term government bonds, +0.03 with long-term government bonds, -0.01 with intermediate-term government bonds.

The correlation between stocks and bonds has been reasonably close to zero. It has not been negative.
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
matjen
Posts: 1694
Joined: Sun Nov 20, 2011 11:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen » Wed Jul 08, 2015 5:38 pm

Yesterdaysnews wrote:Performed poorly today:

Down 1.62% vs. 1.67% for S&P index.


If you are going to compare QSPIX to an equity benchmark even as sort of a laugh, I think VT (Total World) is the only "fair" benchmark. Otherwise you are picking and choosing a subclass of equities.

VT was down 2.11% today.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Mon Jul 13, 2015 6:24 am

nisiprius wrote:
grap0013 wrote:Au contraire mon frère, they are definitely negatively correlated: https://www.portfoliovisualizer.com/ass ... S%2C+VFITX ...
No, they are not.

That's for a short, recent period of time. "Asset correlations for time period 01/28/2011 - 06/29/2015 based on daily returns."

For 1926-2009, inflation-adjusted, the correlation of large-company stocks is +0.23 with long-term government bonds, +0.11 with long-term government bonds, +0.08 for intermediate-term government bond. Nominal, +0.17 with long-term government bonds, +0.03 with long-term government bonds, -0.01 with intermediate-term government bonds.

The correlation between stocks and bonds has been reasonably close to zero. It has not been negative.


Ok, I agree with you. How about this? The historical correlation between stocks and treasury bonds has been low and positive. However, when stocks have bigger drops, treasuries tend to have negative correlation with them. Flight to safety and all that jazz.
There are no guarantees, only probabilities.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Mon Jul 13, 2015 9:06 am

I finally took some time to put my money where my mouth was, pick up a trivial promotional offer for moving a Roth to Fidelity, and dump about 6% of net worth into the fund. Didn't want to use the entire Roth space for it (no rebalancing opportunities), and I got started late on that. I figured the time to get in was before the end of the initial three years, after which Morningstar considers funds for medalists and gives them the kiss of death. 8-)

On a not-that-related side note, I had some morbid curiosity about Vanguard's Alternative Strategies fund, but that doesn't seem to have actually launched (yet?).

randomguy
Posts: 4884
Joined: Wed Sep 17, 2014 9:00 am

Re: QSPIX - thoughts on interesting fund

Post by randomguy » Mon Jul 13, 2015 10:52 am

grap0013 wrote:
Ok, I agree with you. How about this? The historical correlation between stocks and treasury bonds has been low and positive. However, when stocks have bigger drops, treasuries tend to have negative correlation with them. Flight to safety and all that jazz.


The first question is that true? It has been for the past couple (i.e. stocks drops that happened during a bond bull market) but was it also true for 1929-33, 1937, 1962 and 1989 (plus all the ones I can't remember. Seem to remember there was a bad period mid 50s)? And the second of course is does it help you at all. To some extent it comes down to ability to ignore noise. Does it matter if you portfolio drops 30% for 3 months if it closes the year down 10% verus only dropping 20% and closing the year down 9%? Not really.

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Mon Jul 13, 2015 11:17 am

randomguy wrote:
grap0013 wrote:
Ok, I agree with you. How about this? The historical correlation between stocks and treasury bonds has been low and positive. However, when stocks have bigger drops, treasuries tend to have negative correlation with them. Flight to safety and all that jazz.


The first question is that true? It has been for the past couple (i.e. stocks drops that happened during a bond bull market) but was it also true for 1929-33, 1937, 1962 and 1989 (plus all the ones I can't remember. Seem to remember there was a bad period mid 50s)? And the second of course is does it help you at all. To some extent it comes down to ability to ignore noise. Does it matter if you portfolio drops 30% for 3 months if it closes the year down 10% verus only dropping 20% and closing the year down 9%? Not really.

viewtopic.php?t=2409

I wouldn't call it reliable but it does seem to show up significantly.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Mon Jul 13, 2015 2:29 pm

randomguy wrote:Does it matter if you portfolio drops 30% for 3 months if it closes the year down 10% verus only dropping 20% and closing the year down 9%? Not really.


Not to me, but to a lot of people the answer is yes. There is a spectrum of capitulation and everyone has a place on it. Somebodys capitulation point is between a 28% and 30% loss over a 3 month period. Judging by the millions of investors and all the anxiety on here when the market drops a few percent I would say it is a lot of people cumulatively.

We are always talking about mitigating risk on here. One (not you per se) cannot stress the importance of it in one thread and then downplay it in the next. Kinda like fighting over an ER of 0.06% vs. 0.05% and saying how important it is. Then in the next thread said poster says, "I don't care if my 20% bonds cause a 0.5% portfolio drag. It's just 0.5%" It's either important all the time or it isn't.
There are no guarantees, only probabilities.

randomguy
Posts: 4884
Joined: Wed Sep 17, 2014 9:00 am

Re: QSPIX - thoughts on interesting fund

Post by randomguy » Mon Jul 13, 2015 3:56 pm

grap0013 wrote:
randomguy wrote:Does it matter if you portfolio drops 30% for 3 months if it closes the year down 10% verus only dropping 20% and closing the year down 9%? Not really.


Not to me, but to a lot of people the answer is yes. There is a spectrum of capitulation and everyone has a place on it. Somebodys capitulation point is between a 28% and 30% loss over a 3 month period. Judging by the millions of investors and all the anxiety on here when the market drops a few percent I would say it is a lot of people cumulatively.

We are always talking about mitigating risk on here. One (not you per se) cannot stress the importance of it in one thread and then downplay it in the next. Kinda like fighting over an ER of 0.06% vs. 0.05% and saying how important it is. Then in the next thread said poster says, "I don't care if my 20% bonds cause a 0.5% portfolio drag. It's just 0.5%" It's either important all the time or it isn't.


To be fair the difference is that you get nothing for that .01% ER fee while in theory you might be lowering volatiility by holding more bonds. Of course very few of us remember what it is like to invest in bonds when there is like 30 years of flat to rising interest rates. Something like total bond has returned like 7% for the past 30 years to well outpace inflation. You could have been a 100% bond investor the whole time and done ok. I am not sure the 1948 bond investor did anywhere near as well.

Personally I don't think very many people capitulate between 28% and 30%. I expect most of them to have bailed by the time we hit 15%.:)

User avatar
Yesterdaysnews
Posts: 424
Joined: Sun Sep 14, 2014 1:25 pm

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Mon Jul 13, 2015 5:34 pm

The concept of "absolute return" is something I really like. I am not really concerned with the size of returns necessarily rather I like my account to be going up every year regardless of what the market is doing. I think this type of investing can be very appealing, especially to the HNW investor.

If the market was up 15% and my portfolio was up 5% it would bother me less than if the market tanked I my portfolio was down 30% or more with the market.

The risk of the individual investor capitulating I think is underestimated. 2008 was scary for even hardcore disciplined BHeads - just search for threads about "Plan B" from back then .

lack_ey
Posts: 5711
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Mon Jul 13, 2015 5:55 pm

Just don't forget that AQR's fund isn't that much a hedge and includes, among other things, currency carry (recent article by Larry Swedroe).

And absolute return funds, which this doesn't particularly bill itself as (even if in some senses it is) won't that consistently have positive returns. Even taking the optimistic view and assuming the kinds of results seen in AQR's Great Backtest will persist, there will be negative years. To get consistent positive returns, use a savings account, T-bills, and the like.

Out of morbid curiosity, I searched Morningstar for funds with "absolute return" in the name and threw together a bunch of the more established ones (open at least four years and not yet dead) in their charting tool:

Image
Warning: start dates not all the same
http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Tue Jul 14, 2015 7:55 am

randomguy wrote:
Personally I don't think very many people capitulate between 28% and 30%. I expect most of them to have bailed by the time we hit 15%.:)


Ha ha :-)
There are no guarantees, only probabilities.

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

Re: QSPIX - thoughts on interesting fund

Post by matjen » Tue Jul 14, 2015 2:38 pm

lack_ey wrote:I finally took some time to put my money where my mouth was, pick up a trivial promotional offer for moving a Roth to Fidelity, and dump about 6% of net worth into the fund. Didn't want to use the entire Roth space for it (no rebalancing opportunities), and I got started late on that. I figured the time to get in was before the end of the initial three years, after which Morningstar considers funds for medalists and gives them the kiss of death. 8-)

On a not-that-related side note, I had some morbid curiosity about Vanguard's Alternative Strategies fund, but that doesn't seem to have actually launched (yet?).


Welcome to the club lack_ey! Now we just need Robert T.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
Robert T
Posts: 2446
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T » Wed Jul 15, 2015 12:17 am

matjen wrote:
lack_ey wrote:Welcome to the club lack_ey! Now we just need Robert T.

I will sit this one out. I don't need it to achieve my long-term factor load targets. Will stick with the long-only options.

Hope it does well for you. Its an interesting option and will be interesting to see how it performs.

Robert
.

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Wed Jul 15, 2015 9:33 am

^ I think if I could obtain year over year simulated backtested returns I could possibly sway him! :-)

I actually called AQR about 9 months ago and was in discussion with them about it. I got busier at work and it became less of a priority plus I bought 10% anyway. Might still be worth trying to get for those of you on the fence about QSPIX.
There are no guarantees, only probabilities.

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

Re: QSPIX - thoughts on interesting fund

Post by matjen » Wed Jul 15, 2015 9:50 am

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

User avatar
grap0013
Posts: 1867
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Fri Aug 07, 2015 7:21 am

Nothing like a play-by-play update.

Trailing 3 month returns for my portfolio:

Total: -3.42%
QSPIX: +6.31%
VFITX (intermed term treasuries): +0.41%

Those are real returns of a fund that I actually held throughout the entire time frame. Not backtested.
There are no guarantees, only probabilities.

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

Re: QSPIX - thoughts on interesting fund

Post by matjen » Fri Aug 07, 2015 6:03 pm

grap0013 wrote:Nothing like a play-by-play update.

Trailing 3 month returns for my portfolio:

Total: -3.42%
QSPIX: +6.31%
VFITX (intermed term treasuries): +0.41%

Those are real returns of a fund that I actually held throughout the entire time frame. Not backtested.


Phew, I was worried about the Grap Omen but up another .59% today. It has been on quite the run for lackey since he bought some.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
Robert T
Posts: 2446
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T » Fri Aug 07, 2015 6:56 pm

.
:) good for you. I see that all AQRs long/short funds have recently performed relatively well. Over same time frame - past 3 months, and YTD - according to M*.

AQR Equity Market Neutral = 7.76% / 9.59%
AQR Long-Short Equity = 7.70% / 10.67%

Perhaps its momentum thats done relatively well:

iShares MSCI USA Momentum = 4.11% / 7.49%
.

User avatar
Yesterdaysnews
Posts: 424
Joined: Sun Sep 14, 2014 1:25 pm

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Fri Aug 07, 2015 7:45 pm

Active management is not all bad. There are some fund shop which really are very good - AQR is one of them. Another is Artisan funds - really great fund shop. Matthews is great for emerging markets.

I like to use passive index fund for US large cap equities. For other areas, I do believe some top tier active shops add significant value.

rrppve
Posts: 817
Joined: Sun Aug 12, 2012 12:35 am

Re: QSPIX - thoughts on interesting fund

Post by rrppve » Fri Aug 07, 2015 7:57 pm

grap0013 wrote:Nothing like a play-by-play update.

Trailing 3 month returns for my portfolio:

Total: -3.42%
QSPIX: +6.31%
VFITX (intermed term treasuries): +0.41%

Those are real returns of a fund that I actually held throughout the entire time frame. Not backtested.

While these are real returns and I'm also a fan, you did chose an endpoint where QSPIX was up 0.6% on the day when every other asset class had poorer performance, including bonds, equities were off considerably around the world. I do believe that this may have been the day with greatest outperformance of QSPIX as compared to a balanced portfolio over your relatively short holding period.

Post Reply