QSPIX - thoughts on interesting fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
longinvest
Posts: 2447
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest » Thu Jun 11, 2015 8:46 am

Dear Dave (Random Walker),

I have not seen any reference to Jack Bogle in your post.

Best regards,

longinvest
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VAB/ZRR

pkcrafter
Posts: 11871
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: QSPIX - thoughts on interesting fund

Post by pkcrafter » Thu Jun 11, 2015 8:54 am

Investors have curious way of justifying and rationalizing anything they really want to do. :happy

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

Angst
Posts: 1641
Joined: Sat Jun 09, 2007 11:31 am
Location: St Louis, MO

Re: QSPIX - thoughts on interesting fund

Post by Angst » Thu Jun 11, 2015 9:03 am

Robert T wrote:.
It has performed well so far - we have 33 months (almost 3 years) of live returns. See pg 58 of link for returns from Sept. 2012 to Dec 2013. https://www.azasrs.gov/sites/default/fi ... -18-14.pdf - if we add to this the QSPIX returns we get:

Sept 2012 - May 2015 (33 months). Annualized returns (%)

AQR Style Premia = 12.7%
Own portfolio = 12.5%*
Mkt portfolio = 11.1%**

*75:25 Global Stock (with value and small cap tilt):bond
**75:25 Global stock:US total bond (VT:VBMFX)

Robert.

Hi Robert.

I routinely go to M* to compare the growth of various funds and ETFs, but as far as QSPIX goes, M* does not have any data prior to November 2013. The PDF you refer to above is from AQR and shows 2012 "composite assets" for QSPIX amounting to less than $7 million, and the year 2013 for which AQR shows almost 30% gross return, M* only considers performance for the last two months. Incidentally, end of year 2013 assets jump from $7 M to over $1 billion! Presumably, all of this has something to do with why M* growth stats for QSPIX vs. VTI are not nearly as impressive as yours. How do you reconcile it? Is it not prudent to discount or perhaps even reject growth stats provided by AQR for a period of time during which the fund might not have been investable for the likes of us?

Image

Angst
Posts: 1641
Joined: Sat Jun 09, 2007 11:31 am
Location: St Louis, MO

Re: QSPIX - thoughts on interesting fund

Post by Angst » Thu Jun 11, 2015 9:23 am

To grap and Random Walker,

Thank you for posting your % commitments to QSPIX. I'm still a little unclear as to how you decided how much to invest. If the rationale for adding QSPIX is to increase one's portfolio Sharpe ratio, how much QSPIX do you add? If just a little QSPIX (say 10%) is all that is needed, is that to say that your overall portfolio's Sharpe ratio has now been maximized, and that adding more QSPIX would bring it back down? Or are you saying this is the point beyond which overall portfolio returns would begin to fall, volatility not withstanding? Otherwise, one would seem to be falling into a trap that says you should just be 100% QSPIX, which is silly.

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Thu Jun 11, 2015 9:44 am

Sharpe ratio ex ante depends on what you think expected returns, standard deviations, correlations, etc. are going to be and is impossible to figure out with any level of confidence.

Also, maximizing Sharpe with stocks and bonds would look something not far from 20% stocks, 80% bonds. Most people would go for higher risk and higher return at the expense of Sharpe. Perhaps with this, something in general closer to 10% stocks, 50% style premia, 40% bonds, depending hugely upon the input assumptions. Recall that Sharpe is very big on uncorrelated assets because how adding them together works in terms of composite volatility.

I mean, if you expect 7% return here, that might be higher than many forecasts for stocks, with lower volatility. Clearly from MPT this implies a heavier weighting than for stocks. Even if you assume bondlike returns (2% or so) with the 10% standard deviation, you would want some. Less than your bond allocation, but not zero.

Anyway, I think "tracking error" is in general a dumb concern, but I think everyone would caution not to throw huge amounts of money at something unconventional, with more possibility of blowing up or not going as expected. I think most will and should hesitate to put as much as the optimizers or models might suggest.

Random Walker
Posts: 1922
Joined: Fri Feb 23, 2007 8:21 pm

Re: QSPIX - thoughts on interesting fund

Post by Random Walker » Thu Jun 11, 2015 10:07 am

Hi Angst,
I believe adding QSPIX will increase my sharpe ratio from where it is, but I have no idea what the maximum attainable sharpe ratio is or how to get there. My 3% allocation comes from individual portfolio constraints. I am about 85% taxable and 15% tax deferred. I have CCFs, REITs, Int REITs, a very small amount of TIPs, and QSPIX in the tax deferred space and virtually all my bonds are now Municipals in my taxable account. So my allocations to the esoteric tax-inefficient stuff are all very small (but very fun). If I had the room, I'd probably choose a number in the 5-10% range myself. But that is not based on any solid foundation, just my thought.

Dave

User avatar
HomerJ
Posts: 10160
Joined: Fri Jun 06, 2008 12:50 pm

Re: QSPIX - thoughts on interesting fund

Post by HomerJ » Thu Jun 11, 2015 10:54 am

grap0013 wrote:AQR said what the fund would've done in the past and now it's doing it in real life and people still don't like it. This fund could be around in 20 years performing the same way and some people will still be skeptical.


You really think we're crazy for not immediately buying into a new fund's hype?

If it performs through a couple of bear and bull markets over 20 years, then you may have a case...

You really think that because they can show it backtests well, AND it's been performing well for a year or two, that should be proof enough for everyone?

I could list 10,000 funds that had a strategy that back-tested well, performed well for a year or two, and went on to severely under-perform.

Why are we crazy to be skeptical?

User avatar
matjen
Posts: 1671
Joined: Sun Nov 20, 2011 11:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen » Thu Jun 11, 2015 11:01 am

HomerJ wrote:
grap0013 wrote:AQR said what the fund would've done in the past and now it's doing it in real life and people still don't like it. This fund could be around in 20 years performing the same way and some people will still be skeptical.


You really think we're crazy for not immediately buying into a new fund's hype?

If it performs through a couple of bear and bull markets over 20 years, then you may have a case...

You really think that because they can show it backtests well, AND it's been performing well for a year or two, that should be proof enough for everyone?

I could list 10,000 funds that had a strategy that back-tested well, performed well for a year or two, and went on to severely under-perform.

Why are we crazy to be skeptical?


Bzzzt...the answer to Graps question is that people like HomerJ and Longinvest will see the good performance for 20 years (if it happens of course) and then complain that those who desire to hop on then are chasing performance and it is about to mean revert or be grazed away. Just as these threads on long only factor tilts go.
Last edited by matjen on Thu Jun 11, 2015 11:09 am, edited 1 time in total.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
Taylor Larimore
Advisory Board
Posts: 25863
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"Predicting the Past"

Post by Taylor Larimore » Thu Jun 11, 2015 11:05 am

Sharpe ratio ex ante depends on what you think expected returns, standard deviations, correlations, etc. are going to be and is impossible to figure out with any level of confidence.

A very wise Morningstar poster, username "Ozark," agrees:
PREDICTING THE PAST:

If you feel you can improve your portfolio's asset allocation by running the portfolio through various computer programs, measuring and grading various risk/reward relationships, feel free. It's okay with me. Honest. For myself, I'm not interested.

I'm also not interested in running reams of data through a computer program in order to discover how much I can withdraw yearly from my portfolio and never go broke.

Without having studied it, I'm willing to assume the Risk Grades deal is similar to the well known Efficient Frontier concept: Invest in a mix of assets that will give the best return for the least risk.

Wonderful. The problem in execution is this; both these approaches would seem to be limited to looking at PAST risk/return relationships, in order to predict FUTURE such relationships.

This approach hasn't worked very well and it never will.

There's lots of stuff we can learn by studying the past. One thing we can't learn, though, is how much the future will resemble the past.

There really is an Efficient Frontier. There really is a withdrawal rate that will allow my wife and I to spend all our money during our life times, but never go broke.

But these things are unknown and unknowable, going forward. Such things are only knowable looking backward.

Given that such things are only knowable looking backward, academics with more letters after their names than I have money in the bank, have spent unconscionable amounts of time goobering through the past. They thus invented Modern Portfolio Theory---Beta, Alpha, R-Squared, and the crowning achievement, Sharpe Ratio. These accomplishments were celebrated and awards were given. Yes.

And then...a funny thing happened on the way to the bank. These numbers turned out to have little or no predictive value, regarding returns. And since they couldn't predict returns, they also failed to predict risk/return ratios.

Joining in the fun, M* invented their first Star Rating system, a system that graded...yep...risk- adjusted, past performance.

I wish I had 10 bucks for every post I've read where the poster said, essentially, "I have a balanced portfolio, made up entirely of 4 and 5 star funds." Too late, these jokers discovered what M* eventually discovered; past risk-adjusted performance doesn't predict future risk-adjusted performance.

I don't want to discover the Sharpe Ratio of my portfolio. I don't want to discover its Beta. I don't want to discover its Risk Grade. I have absolutely no confidence that adjusting the portfolio so that these numbers become more favorable will improve future risk/reward.

If others do want to do that, that's okay with me. I seriously doubt, though, that many successful mutual fund managers select securities in that manner. If any do, or if any money managers set their asset allocations in that manner, I'd be interested in their long-term results---results over periods of, say, 10 years, or more.

In short, computers are wonderous tools, but that's all they are. Every computer on Earth, all linked up and working 24/7, from now on, won't tell me my survivable withdrawal rate. Neither will they tell me what asset allocation would give me the best risk/reward ratio.

In my opinion, these things can't be calculated. We have to forge ahead without knowing these things. Deal with it.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Thu Jun 11, 2015 11:29 am

HomerJ wrote:
grap0013 wrote:AQR said what the fund would've done in the past and now it's doing it in real life and people still don't like it. This fund could be around in 20 years performing the same way and some people will still be skeptical.


You really think we're crazy for not immediately buying into a new fund's hype?

If it performs through a couple of bear and bull markets over 20 years, then you may have a case...

You really think that because they can show it backtests well, AND it's been performing well for a year or two, that should be proof enough for everyone?

I could list 10,000 funds that had a strategy that back-tested well, performed well for a year or two, and went on to severely under-perform.

Why are we crazy to be skeptical?

Any and all investing has to be based partially on historical returns. Go back enough centuries and you don't see much or any outperformance by equities.

The underlying (claimed) sources of return for the fund are based on strategies that are largely not new and have a historical track record, with all of the warts that entails. Some, particularly in equities, go way back before the cited 23-year backtest.

You should always be skeptical, but if you can't address these underlying strategies with more specifics then we're going nowhere here.

User avatar
grap0013
Posts: 1861
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Thu Jun 11, 2015 1:10 pm

longinvest wrote:Food for thought:
  • For QSPIX to succeed, it needs dumb investors on the other side of every transaction. Why dumb? Because if QSPIX trades using solid scientific strategies supposed to always win, then these traders should be guaranteed to loose. They must really be dumb to doubt science and continue betting against QSPIX.
  • These dumb investors cannot be total-market index investors who do not trade within the measured period T (see my previous post).
  • Retail investors are now a minority in markets. QSPIX can't just rely on them to make a killing. So, can you identify who the dumb investors are? If not: Look around the Poker table; if you can’t see the sucker, you’re it.


People must be really dumb to doubt science and buy bonds instead of equities. Do you see the fault in your above logic? You are forgetting the aspect of risk. I'm not saying I think QSPIX is guaranteed to be awesome. I do think it is highly likely to be helpful in the context of an addition to a diversified portfolio. You have to put your money somewhere.
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1861
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Thu Jun 11, 2015 1:26 pm

HomerJ wrote:
grap0013 wrote:AQR said what the fund would've done in the past and now it's doing it in real life and people still don't like it. This fund could be around in 20 years performing the same way and some people will still be skeptical.


You really think we're crazy for not immediately buying into a new fund's hype?

If it performs through a couple of bear and bull markets over 20 years, then you may have a case...

You really think that because they can show it backtests well, AND it's been performing well for a year or two, that should be proof enough for everyone?

I could list 10,000 funds that had a strategy that back-tested well, performed well for a year or two, and went on to severely under-perform.

Why are we crazy to be skeptical?


Nope. I never said crazy and never said proof enough for everyone. I will say, however, some Bogleheads are 3+ standard deviations outside the norm on the Guassian curve of skepticism. We are all unique. I think some skepticism is definitely good, but sometimes it's an unhealthy amount. How do some people so doubting make decisions in their everyday life? Are airplanes really safer than cars? Which should I take? We only have 100 years of data. Is that enough to know for sure? It borders on conspiracy theorist territory sometimes. I am not specifically point out to you Paul. Just making a general statement.

We rarely have certainty in life and I just stack my poker chips where I think I have the best odds for success. To each their own.

BTW, I'm not saying this is all related to this fund QSPIX. I understand more skepticism with it due to its newness and complexity. But for instance, SCV has been around for 23 years even in out of sample tests and it's statistically significant. The premiums are found in international stocks as well, but many still ignore the facts.

If posters are going to be skeptical about QSPIX at least look at the data a little for yourself first rather than just jerking your knees with a canned "it won't work" response. That's all I'm asking. Some people actually want to talk about the nuts and bolts of the fund rather than every other post being, "look at that ER", "it's garbage" "blah blah blah". I don't post in all the TSM and 3 fund threads about how everyone should tilt to SCV. See my point?

Typed words can be taken out of context and I mean all of the above with the utmost respect and I am not trying to offend anyone.

Thanks!

grap
There are no guarantees, only probabilities.

wickywack
Posts: 37
Joined: Thu Jun 11, 2015 8:09 am

Re: QSPIX - thoughts on interesting fund

Post by wickywack » Thu Jun 11, 2015 1:27 pm

I've been reading Bill Bernstein's "Skating Where the Puck Was: The Correlation Game in a Flat World". His main theme is that financial history shows over and over again that risky assets that haven't correlated with the market at the beginning usually start to correlate strongly once they get popular.

Wouldn't that potentially apply to this fund as well?

From his book:

Remember, investing early in true alternative asset classes is, by definition, never easy, and once they do become securitized, liquid, and popular, all bets are off. In the digital age, any risky asset class you can buy with a keystroke can, and most likely, will bite you when things head south.


and

To summarize, throughout all of financial history, it has been difficult, if not impossible, to invest in risky asset classes that are both highly liquid and short-term non-correlating, since liquidity, as we've seen, jacks up short-term correlations.

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

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Thu Jun 11, 2015 2:25 pm

If I bought this fund in a traditional IRA account in Fidelity could I transfer it my traditional IRA account at Vanguard? This would be a nice fund to hold at Vanguard and mix with admiral shares of Total Stock market. If they say no they would lose my taxable account also.

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

Re: QSPIX - thoughts on interesting fund

Post by countmein » Thu Jun 11, 2015 2:26 pm

wickywack wrote:I've been reading Bill Bernstein's "Skating Where the Puck Was: The Correlation Game in a Flat World". His main theme is that financial history shows over and over again that risky assets that haven't correlated with the market at the beginning usually start to correlate strongly once they get popular.

Wouldn't that potentially apply to this fund as well?


QSPIX is different in that it is a bunch of trading strategies, not assets. So the anti-correlation shouldn't go away but the premia from the strategies could (should?) evaporate with increased popularity/availability.

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Thu Jun 11, 2015 2:28 pm

grap0013 wrote:
longinvest wrote:Food for thought:
  • For QSPIX to succeed, it needs dumb investors on the other side of every transaction. Why dumb? Because if QSPIX trades using solid scientific strategies supposed to always win, then these traders should be guaranteed to loose. They must really be dumb to doubt science and continue betting against QSPIX.
  • These dumb investors cannot be total-market index investors who do not trade within the measured period T (see my previous post).
  • Retail investors are now a minority in markets. QSPIX can't just rely on them to make a killing. So, can you identify who the dumb investors are? If not: Look around the Poker table; if you can’t see the sucker, you’re it.


People must be really dumb to doubt science and buy bonds instead of equities. Do you see the fault in your above logic? You are forgetting the aspect of risk. I'm not saying I think QSPIX is guaranteed to be awesome. I do think it is highly likely to be helpful in the context of an addition to a diversified portfolio. You have to put your money somewhere.

Okay, a few things here.

1. Stretching a stock/bond example to long/short style premia is going way too far because the context is if somebody is losing on the other side of the trade. A hedged fund like this has no net exposure to the underlying asset classes. Within each asset class, for the fund to make money it needs to net pick assets that do relatively well. It needs to buy underpriced and sell overpriced. This is a very "active" mindset, not one of price taking, for better or worse.

2. All of longinvest's points apply to anybody who deviates from total market investing. Though not substantially off, even those in total market stock and bond index funds are not total market investors. They don't own private equity, never mind some small companies that don't make the index. Depending on how you want to count things, they may be missing commodities, real estate, and all sorts of other assets. Many don't own any or much bonds globally. But even if we restrict thinking to a given market such as the universe of stocks included in a total stock market fund, all these points are directed at anybody overweighting REITs or value tilting or who knows what else. Long-only bets are bets against the market too. If someone believes only in total market investing within a given asset class, then this is a position that is fine and very defensible, though I wonder why such a person is not repeating these things in every other thread where anyone considers any kind of tilt.

3. For longinvest's first point, it is a ridiculous to suggest that anybody is talking about a strategy that will "always win." This isn't a pure arbitrage by any stretch of the imagination. The super-datamined-fantasyland-backtest itself for the styles includes a drawdown of 15%. The component pieces are each just supposed to work more often than not and make money in the long run. It's perhaps not much different from the equity risk premium in this respect: usually positive in the long run, negative in shorter periods a lot. If you think the equity risk premium is likely to continue to be positive while these things are not likely to be positive (or not positive enough to cover costs and fees), then that is plenty fine. Just say it straight, and give all the reasons why, if you could, so we can all understand better.

4. For longinvest's third point, every group of investors could be contributing to the dumbness, if we want to call it that, to different degrees. That said, here are some examples of where this dumbness may come from. Even if mutual funds are smart, if investors market time their way in and out of them as still seems to be the case, the net effect may be (for example) giving up on value stocks after value has done poorly, from the pattern of redemptions. Carry and value could persist in bonds and carry in currencies and short-term rates if the "dumb" investors or entities controlling the markets are governments setting policy, insurance companies with nominal rather than real liabilities willing to go for bonds we wouldn't, and a variety of companies and institutions required to buy and hold certain assets by policies and regulations. Defensive might work in equities because many institutional investors are restricted (or prefer) not to use any leverage, shorting, or derivatives.

That said, I agree a lot of humility is required and thinking before embarking on a quest to beat the market, even if you think it is based on good evidence. I'm also one of many who suggest that picking individual securities is very hard because you're playing against institutional money. It just may or may not be impossible.

wickywack wrote:I've been reading Bill Bernstein's "Skating Where the Puck Was: The Correlation Game in a Flat World". His main theme is that financial history shows over and over again that risky assets that haven't correlated with the market at the beginning usually start to correlate strongly once they get popular.

Wouldn't that potentially apply to this fund as well?

From his book:

Remember, investing early in true alternative asset classes is, by definition, never easy, and once they do become securitized, liquid, and popular, all bets are off. In the digital age, any risky asset class you can buy with a keystroke can, and most likely, will bite you when things head south.


and

To summarize, throughout all of financial history, it has been difficult, if not impossible, to invest in risky asset classes that are both highly liquid and short-term non-correlating, since liquidity, as we've seen, jacks up short-term correlations.

I think this line of thinking is worth pursuing and contemplating, but remember that this is not investing in any kind of newly available risky assets. It is net zero invested in assets, being hedged long/short, intentionally restricting itself to the most liquid securities. It doesn't even allow itself to use stocks that are too small.

I mean, currencies ain't exactly new. And you could definitely have invested in say copper futures or some of the more liquid Japanese or British equities or German bunds back in 1980, which is where the backtest starts (good data was not available for some asset classes used before then).

Random Walker
Posts: 1922
Joined: Fri Feb 23, 2007 8:21 pm

Re: QSPIX - thoughts on interesting fund

Post by Random Walker » Thu Jun 11, 2015 3:51 pm

Wickywack,
I agree with you that Skating Where The Puck Was is important reading. I also agree that when an excellent weakly correlated asset class becomes liquid and investable, future expected returns can be muted. And I agree that when crises occur, all correlations head towards one. But diversification benefits from weak correlation can occur when the markets are not in crisis.

Dave

User avatar
TdF fan
Posts: 119
Joined: Sat Mar 01, 2008 2:35 pm

Re: QSPIX - thoughts on interesting fund

Post by TdF fan » Thu Jun 11, 2015 5:38 pm

For those of you who have an allocation to this fund, I'm curious if you plan to rebalance into or out of it as applicable (with whatever rules you normally use to rebalance)?

I'm just trying to figure out if this fund represent an asset class that would be rebalanced similar to stocks and bonds, or is it a different animal with different rules.

Random Walker
Posts: 1922
Joined: Fri Feb 23, 2007 8:21 pm

Re: QSPIX - thoughts on interesting fund

Post by Random Walker » Thu Jun 11, 2015 6:50 pm

I have 3% allocation and will rebalance with it same as any other asset in the portfolio.

Dave

robert88
Posts: 366
Joined: Tue Nov 25, 2014 6:27 pm

Re: QSPIX - thoughts on interesting fund

Post by robert88 » Thu Jun 11, 2015 7:58 pm

grap0013 wrote:Nope. I never said crazy and never said proof enough for everyone. I will say, however, some Bogleheads are 3+ standard deviations outside the norm on the Guassian curve of skepticism. We are all unique. I think some skepticism is definitely good, but sometimes it's an unhealthy amount. How do some people so doubting make decisions in their everyday life? Are airplanes really safer than cars? Which should I take?
grap


I don't think anyone makes the decision to drive vs. fly based on safety, but let's assume they do and I think you'll see what's wrong with the tilting approach. The factor tilters would say from 2000-2013, deaths from cars per 100 million miles averaged about 1.3 and deaths from commercial air travel, excluding terrorism, were almost zero. Therefore, the quants would conclude that flying will always be safer than driving for hundreds of years into the future. Someone taking an approach based more on fundamentals, would see that self driving cars are inevitable and that after the initial bugs get resolved, the fatality rate from auto accidents will eventually be near zero. It's a question of when, not if, that "factor" gets "grazed away", but you wouldn't know that if you based all your decisions based on historical, long run averages.

Fryxell
Posts: 66
Joined: Sat Mar 02, 2013 10:08 pm
Location: Dallas

Re: QSPIX - thoughts on interesting fund

Post by Fryxell » Thu Jun 11, 2015 8:14 pm

robert88 wrote:It's a question of when, not if, that "factor" gets "grazed away", but you wouldn't know that if you based all your decisions based on historical, long run averages.


Do you think the equity premium will be grazed away? The idea of the equity premium is also based on historical averages.

I find it peculiar how people that don't believe in factor tilts nevertheless believe in tilting towards equities.

robert88
Posts: 366
Joined: Tue Nov 25, 2014 6:27 pm

Re: QSPIX - thoughts on interesting fund

Post by robert88 » Thu Jun 11, 2015 8:21 pm

Fryxell wrote:
robert88 wrote:It's a question of when, not if, that "factor" gets "grazed away", but you wouldn't know that if you based all your decisions based on historical, long run averages.


Do you think the equity premium will be grazed away? The idea of the equity premium is also based on historical averages.

I find it peculiar how people that don't believe in factor tilts nevertheless believe in tilting towards equities.


I think stocks are clearly riskier than bonds(junk bonds and exotic instruments like CDOs excepted) and that as long as investors on average are at least somewhat rational, they will demand a "risk premium" to compensate them for the additional risk that stocks have over bonds. I don't have to look at any historical data to reach that conclusion.

Fryxell
Posts: 66
Joined: Sat Mar 02, 2013 10:08 pm
Location: Dallas

Re: QSPIX - thoughts on interesting fund

Post by Fryxell » Thu Jun 11, 2015 10:07 pm

robert88 wrote:
I think stocks are clearly riskier than bonds(junk bonds and exotic instruments like CDOs excepted) and that as long as investors on average are at least somewhat rational, they will demand a "risk premium" to compensate them for the additional risk that stocks have over bonds. I don't have to look at any historical data to reach that conclusion.


Well, small stocks are clearly riskier than large stocks. Distressed (often value) companies are clearly riskier than non-distressed companies. So shouldn't they also have a risk premium, by the same logic?
Last edited by Fryxell on Thu Jun 11, 2015 11:25 pm, edited 1 time in total.

robert88
Posts: 366
Joined: Tue Nov 25, 2014 6:27 pm

Re: QSPIX - thoughts on interesting fund

Post by robert88 » Thu Jun 11, 2015 11:04 pm

Fryxell wrote:Well, small stocks are clearly riskier than large stocks. Distressed (often value) companies are clearly riskier than non-distressed companies. So shouldn't they also have a risk premium, by the same logic?


Maybe, there's actually a lot of debate on here about whether small and value really are riskier and if so, how much their risk can explain their premiums. According to AQR, defensive = low risk stocks(possibly leveraged) - high risk stocks, as they explain it at as the tendency for lower-risk and higher-quality assets to generate higher risk-adjusted returns. If you explain away value as high risk stocks - low risk stocks, then it would appear loading on both defensive and value should nearly cancel.

That said, if you're like Larry Swedroe and only put 3% of your portfolio into QSPIX, that's such a small percentage it's effect on your portfolio's overall return will probably be beneath the noise floor, regardless of how it turns out.

Fryxell
Posts: 66
Joined: Sat Mar 02, 2013 10:08 pm
Location: Dallas

Re: QSPIX - thoughts on interesting fund

Post by Fryxell » Thu Jun 11, 2015 11:31 pm

robert88 wrote:Maybe, there's actually a lot of debate on here about whether small and value really are riskier and if so, how much their risk can explain their premiums. According to AQR, defensive = low risk stocks(possibly leveraged) - high risk stocks, as they explain it at as the tendency for lower-risk and higher-quality assets to generate higher risk-adjusted returns. If you explain away value as high risk stocks - low risk stocks, then it would appear loading on both defensive and value should nearly cancel.

That said, if you're like Larry Swedroe and only put 3% of your portfolio into QSPIX, that's such a small percentage it's effect on your portfolio's overall return will probably be beneath the noise floor, regardless of how it turns out.


Well, I don't believe that premiums are entirely because of risk. I think some are behavioral. I agree with Swedroe that behavioral premiums can persist because of irrationality and limits to arbitrage. And as you noted, some risks are not compensated, such as those in junk bonds. My point is rather that it is curious how nobody seems to doubt the equity premium, even though the evidence for it is exactly the same as for the value and size premiums (historical and risk based); and in the case of momentum the evidence is that it is stronger than the equity premium. And furthermore, stocks underperformed bonds for 40 years from 1969 to 2009, yet no boglehead seems to doubt the existence of the equity premium, including Bogle himself. It is rather curious to me. I think it is because of mental "bucketing" where people mentally categorize stocks and bonds as fundamentally different, thus making it easier for them to rationalize the risk premium for stocks.

The defensive/low volatility is debatable. Some have explained it as a combination of value, quality and term exposure. But if defensive is real, then it would have to be behavioral, rather than risk-based.

I agree that putting 3% of a portfolio in anything is not going to make enough of a difference to be worth bothering with.

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Thu Jun 11, 2015 11:45 pm

Fryxell wrote:
robert88 wrote:Maybe, there's actually a lot of debate on here about whether small and value really are riskier and if so, how much their risk can explain their premiums. According to AQR, defensive = low risk stocks(possibly leveraged) - high risk stocks, as they explain it at as the tendency for lower-risk and higher-quality assets to generate higher risk-adjusted returns. If you explain away value as high risk stocks - low risk stocks, then it would appear loading on both defensive and value should nearly cancel.

That said, if you're like Larry Swedroe and only put 3% of your portfolio into QSPIX, that's such a small percentage it's effect on your portfolio's overall return will probably be beneath the noise floor, regardless of how it turns out.


Well, I don't believe that premiums are entirely because of risk. I think some are behavioral. I agree with Swedroe that behavioral premiums can persist because of irrationality and limits to arbitrage. And as you noted, some risks are not compensated, such as those in junk bonds. My point is rather that it is curious how nobody seems to doubt the equity premium, even though the evidence for it is exactly the same as for the value and size premiums (historical and risk based); and in the case of momentum the evidence is that it is stronger than the equity premium. And furthermore, stocks underperformed bonds for 40 years from 1969 to 2009, yet no boglehead seems to doubt the existence of the equity premium, including Bogle himself. It is rather curious to me. I think it is because of mental "bucketing" where people mentally categorize stocks and bonds as fundamentally different, thus making it easier for them to rationalize the risk premium for stocks.

The defensive/low volatility is debatable. Some have explained it as a combination of value, quality and term exposure. But if defensive is real, then it would have to be behavioral, rather than risk-based.

I agree that putting 3% of a portfolio in anything is not going to make enough of a difference to be worth bothering with.

They mean defensive in the sense of low beta and don't completely ascribe the value premium to risk (they would say it's a combination of risk and behavioral and other mispricing). There's a clear distinction between defensive and value, and they are much closer to orthogonal than opposite.

Low beta, if you accept this as is, is behavioral, but some explanations for it would explain that behavior as rational on the part of money managers—say, following from constraints on leverage and short selling imposed on many market players. Most of the times people describe anomalies as behavioral, they mean there's a lapse in judgment and a mistake, which is not necessarily the case here.

Anyway, the equity risk premium is considered by many to have the most robust theoretical underpinnings, though (spotty) data prior to the 20th century doesn't show much evidence of it, so it may be a more modern phenomenon. Usually this is meant relative to the risk-free rate; bonds doing better than stocks over a lengthy period in the US is only if you look at longer-term bonds and that is just from the term risk premium being huge for much of the time, on top of rapidly falling inflation and rates.

User avatar
nedsaid
Posts: 8286
Joined: Fri Nov 23, 2012 12:33 pm

Re: QSPIX - thoughts on interesting fund

Post by nedsaid » Fri Jun 12, 2015 12:45 am

Would I invest in such a fund? It depends.

If I were constructing a portfolio on my own, which I am doing, I would not. Were I employing someone like Larry Swedroe as my advisor, I would probably own this as part of the package. But I would have to have someone break this down into its component parts and explain each part to me.

I have a natural skepticism towards quantitative strategies because when you boil it down, you are attempting to establish precise measurements to human behavior. Human behavior is the wild card.

You take any equation you want and multiply it by H, H being the human factor. H has a value of zero. It is an oversimplification, but it does illustrate the unpredictability of human behavior. Throw H in there and it can destroy the best constructed models and formulas.

This is why the performance of asset classes, the relationship in performance of asset classes to each other, and the volatility of asset classes change all the time. It is those darned humans and their emotions, primarily fear and greed, that screw all of this up.

To the extent that factor investing works because of human nature and human behavior, and that over time humans perform the same behavioral errors over and over, I am with you. If you have enough patience and a long enough time horizon, you should have enough time for market cycles to play out and for human behavioral errors to repeat.

Where you lose me is when the quants try to precisely measure the effects of human behavior. The mathematical relationships calculated from history are not going to replicate themselves exactly into the future. In fact there might be a big difference between past relationships and the future. In other words, history doesn't repeat itself, it just rhymes. For example, we know that a bear market is coming. What we don't know is what will cause it. In fact, each bear market is different and each has somewhat different causes.

The problem with this is the old fighting the last war problem. What worked in past markets may not work in future markets. Trying to guard against Guderian's Panzer Divisions in 1940 with the Maginot Line. The Maginot Line would have been brilliant in WW I but of limited utility in WWII.

I also operate under the theory that the more moving parts you have, the bigger odds of a breakdown. This fund has a lot of moving parts and there are more things that can go wrong.

I never say never. If I really buckled down and really studied this fund to where I could understand what it was doing, I might consider it. The fund does things like shorting and investing in futures that I have no expertise or experience in. There are a few Bogleheads who have experience in these areas and would feel comfortable in such a fund. I plead inexperience and ignorance and choose to pass for now.
A fool and his money are good for business.

User avatar
grap0013
Posts: 1861
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Fri Jun 12, 2015 6:31 am

3Factor wrote:For those of you who have an allocation to this fund, I'm curious if you plan to rebalance into or out of it as applicable (with whatever rules you normally use to rebalance)?

I'm just trying to figure out if this fund represent an asset class that would be rebalanced similar to stocks and bonds, or is it a different animal with different rules.


I put 10% into QSPIX. I know the general rule is a 5% allocation to make it worth your while but 10% is my personal minimum. Normal rebalancing bands around this guy. It's not rebalancing into a black hole (eg it's not going to $0).
There are no guarantees, only probabilities.

Random Walker
Posts: 1922
Joined: Fri Feb 23, 2007 8:21 pm

Re: QSPIX - thoughts on interesting fund

Post by Random Walker » Fri Jun 12, 2015 8:50 am

Grap0013,
I think it's worthwhile to note that for really volatile or weakly/negatively correlated asset classes, their diversification effect on a portfolio will appear to be bigger than their small % allocation would suggest. Don't remember how to calculate it, but I've seen that sort of example with CCFs.

Dave

User avatar
TdF fan
Posts: 119
Joined: Sat Mar 01, 2008 2:35 pm

Re: QSPIX - thoughts on interesting fund

Post by TdF fan » Fri Jun 12, 2015 9:07 am

grap0013 wrote:
3Factor wrote:For those of you who have an allocation to this fund, I'm curious if you plan to rebalance into or out of it as applicable (with whatever rules you normally use to rebalance)?

I'm just trying to figure out if this fund represent an asset class that would be rebalanced similar to stocks and bonds, or is it a different animal with different rules.


I put 10% into QSPIX. I know the general rule is a 5% allocation to make it worth your while but 10% is my personal minimum. Normal rebalancing bands around this guy. It's not rebalancing into a black hole (eg it's not going to $0).

Thanks grap and RW.

After reading the Forbes article (link below) I understand the general idea better now, but I find this quote from Antti Ilmanen to be telling:
Antti Ilmanen wrote:Practical implementation is an area to which I had not paid sufficient attention before I joined AQR. Especially in a world of low return premia one cannot afford to be wasteful; there may be net gains only for the most cost-effective harvesters.

http://www.forbes.com/sites/phildemuth/2013/12/09/the-best-investment-book-of-the-past-ten-years-is-now-a-mutual-fund-plus-q-a-with-antti-ilmanen-and-ronen-israel/2/

User avatar
Ketawa
Posts: 1840
Joined: Mon Aug 22, 2011 1:11 am
Location: Norfolk

Re: QSPIX - thoughts on interesting fund

Post by Ketawa » Fri Jun 12, 2015 9:09 am

I rebalance as normal with the caveat that I use QSPNX until it makes sense for me to pay $50 to dump more into QSPIX rather than paying an ER that is 25 bp higher.

User avatar
matjen
Posts: 1671
Joined: Sun Nov 20, 2011 11:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen » Fri Jun 12, 2015 9:19 am

grap0013 wrote:If posters are going to be skeptical about QSPIX at least look at the data a little for yourself first rather than just jerking your knees with a canned "it won't work" response. That's all I'm asking. Some people actually want to talk about the nuts and bolts of the fund rather than every other post being, "look at that ER", "it's garbage" "blah blah blah". I don't post in all the TSM and 3 fund threads about how everyone should tilt to SCV. See my point?

Typed words can be taken out of context and I mean all of the above with the utmost respect and I am not trying to offend anyone.

Thanks!

grap


Don't worry grap. I'm in charge of being snarky in defense of QSPIX and factor-based investing. :D

I have been reposting something Robert T. put up a couple weeks back because it struck me as so impressive and wasn't back-tested. His portfolio since he has had it compared to the equivalent Vanguard LifeStrategy (AKA the standard 3-fund for all practical purposes). I love the standard 3-fund and LifeStrategy is my favorite single fund I suppose. This is the strategy I discuss with any friend who wants to know more about investing but doesn't want to make it a hobby. I own it in my HSA. That doesn't mean that factor-based investing isn't an incredible alternative for many. Actually based on polls on this site the majority. It seems to me that people with 3 fund in their sig line and such should show a little more humility to other "similar" strategies. We are all members of the same "church." You and I may prefer just a different 3-fund if we were starting from scratch. Perhaps 3.1 DFA VFVEX for US, DFA DWUSX for international, and Schwab SCHR for bonds. Or, more relevant to this thread, 3.2 QSPIX, Vanguard VT, and Schwab SCHR. Just spitballing here.

Robert T wrote:
.
Out of interest I just checked my portfolio performance since inception (at start of 2003) against the Vanguard lifestrategy growth funds. I have a 75:25 stock:bond portfolio with global small cap and value tilt. A combined 75:25 Vanguard Lifestrategy Growth:Moderate Growth mix provides a 75% stock:25% bond comparison.

Over last 12 years: Annualized return (%)
Lifestrategy fund combination = 8.3% (2008 return = -32.4%)
Own portfolio = 10.3% (2008 return = -28.7)

The 2008 downside would have been matched with a 35% bond allocation among the Lifestrategy fund combination for an annualized return over the last 12 years of 7.8%. In addition, using the lifestrategy funds would likely have been less tax efficient, as would need to hold some bonds in taxable account.

Obviously no guarantees.

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

User avatar
Ketawa
Posts: 1840
Joined: Mon Aug 22, 2011 1:11 am
Location: Norfolk

Re: QSPIX - thoughts on interesting fund

Post by Ketawa » Fri Jun 12, 2015 9:32 am

Are those LifeStrategy returns based on simulated returns or the actual returns of the funds? They used to have an allocation to a Vanguard tactical asset allocation fund which timed the market poorly.

User avatar
matjen
Posts: 1671
Joined: Sun Nov 20, 2011 11:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen » Fri Jun 12, 2015 9:58 am

Ketawa wrote:Are those LifeStrategy returns based on simulated returns or the actual returns of the funds? They used to have an allocation to a Vanguard tactical asset allocation fund which timed the market poorly.


Excellent point. Not sure what Robert T. did but it was in response to this thread which also included the LifeStrategy 10 yr returns.

viewtopic.php?f=1&t=166449&p=2502131#p2502131

So here's the lesson. And what's caused me to pause and reflect.

Vanguard's Lifestrategy Conservative Growth Fund (60% bonds/40% stocks) has earned:

6.94% -- 1 year; 7.14% -- 5 years; 5.75% -- 10 years.

Its cost is 15 basis points.

So all my slicing and dicing? All my analysis? And tweaking? And fretting and studying and second-guessing myself every other Thursday? It's brought me to within a quarter percent of Lifestrategy Conservative Growth after ten freaking years, and (saddest of all) a quarter percent lower.

And so, ladies and gents, maybe just maybe Taylor Larimore and Jack Bogle have some salient points. Maybe there is "Majesty in Simplicity." And also extra return.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
nisiprius
Advisory Board
Posts: 33794
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 12, 2015 10:11 am

Random Walker wrote:Grap0013,
I think it's worthwhile to note that for really volatile or weakly/negatively correlated asset classes, their diversification effect on a portfolio will appear to be bigger than their small % allocation would suggest. Don't remember how to calculate it, but I've seen that sort of example with CCFs.

Dave
I do not think this is correct, but it's hard to address it without specifics and without knowing what "bigger than their % allocation would suggest" means.

I agree, though, that in my informal explorations, correlations of 0.7 (e.g. international and U.S., small-cap and total market) are much less effective than people seem to think, unless you believe that all theoretical improvements, no matter how small, must matter. It seems to take correlations of 0.25 or less to start getting interesting. The devil of it, of course, is that to be interesting they have to be robust and persistent, and in real life that doesn't seem to be common--low correlations often seem to be sampling error, the result of a short period of time.

Everything's relative, but if I look at a mixture of Vanguard Total Stock and PIMCO Commodity Real Return Strategy, this is what I see. If I assume that I want my total portfolio standard deviation to be 10%, then I get an improvement of 0.66% but in order to get it, I need to commit to a full 35% allocation to the commodities fund. I'm too lazy to get the exact number but to a rough approximation, I might expect a 10% allocation to give me an improvement of 0.2%, though because of the curve of the line it is actually going to be less than that.

So, that's an 0.20% improvement in improvement, for a 10% allocation--in hindsight, over that particular past period.

The correlation over that time period was low, only 0.165. However, after reading "skating where the puck was," one has to wonder about persistence. It does not seem to me that it makes sense to apply a fancy strategy that theoretically gives you an 0.20% improvement unless you have already squeezed your weighted-average portfolio expense ratio down into the same neighborhood.

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.

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: QSPIX - thoughts on interesting fund

Post by Brogleski » Fri Jun 12, 2015 10:17 am

matjen wrote: Or, more relevant to this thread, 3.2 QSPIX, Vanguard VT, and Schwab SCHR. Just spitballing here.

Not trying to pick on this quote, but isn't holding VT and QSPIX at the same time inefficient (from a cost perspective)? It seems to me that one would be able to use SCHR+multi-factor equity funds (PRF,PXF,PXH, etc.) to reach the same loadings as a VT+QSPIX+SCHR portfolio with lower fees.
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

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

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews » Fri Jun 12, 2015 10:20 am

I would agree that a small allocation to this fund may not do much, I would say a 30% allocation and you will see some interesting portfolio moves.

User avatar
grap0013
Posts: 1861
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 » Fri Jun 12, 2015 10:26 am

Random Walker wrote:Grap0013,
I think it's worthwhile to note that for really volatile or weakly/negatively correlated asset classes, their diversification effect on a portfolio will appear to be bigger than their small % allocation would suggest. Don't remember how to calculate it, but I've seen that sort of example with CCFs.

Dave


I just run a couple of the back of a napkin calculations. If 97% of my portfolio drops 50% and the other 3% goes up by 10% there is not much effect. Maybe it shows up in the Sharpe Ratio, but does not do much for me in real life. In medical terms, something may be statistically significant, but clinically irrelevant (eg. I reduce the risk of a moderately serious event by 50%, but my baseline event rate is 1 in 200,000). Not dissing your strategy, but for me personally, it has to be 10% or it's not worth messing with.
There are no guarantees, only probabilities.

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 12, 2015 12:09 pm

nedsaid wrote:Would I invest in such a fund? It depends.

If I were constructing a portfolio on my own, which I am doing, I would not. Were I employing someone like Larry Swedroe as my advisor, I would probably own this as part of the package. But I would have to have someone break this down into its component parts and explain each part to me.

I have a natural skepticism towards quantitative strategies because when you boil it down, you are attempting to establish precise measurements to human behavior. Human behavior is the wild card.

You take any equation you want and multiply it by H, H being the human factor. H has a value of zero. It is an oversimplification, but it does illustrate the unpredictability of human behavior. Throw H in there and it can destroy the best constructed models and formulas.[...]

To be honest, I don't think this is really much of a quantitative strategy in the sense that you might think, of requiring predictions or more complicated modeling. If one sorts stocks by price/book and holds the ones with low multiples, do you count that as a quantitative strategy? Here, most of the secret sauce is just in the trading and portfolio management, seeking the least costly ways to transition portfolio holdings and manage leverage slightly up and down as volatility changes. It's about managing and rebalancing a portfolio of many constituent pieces without going overboard and racking up too many costs.

The underlying strategies in each style/asset class combination are not difficult to understand. The styles are value, momentum, carry, and defensive.

Value:
1. Stocks — low price/book
2. Country equity indexes — same as individual stocks
3. Currencies — high real exchange rate (you want the currency that buys more goods/services)
4. Bonds — real 10-year yield (based on current 10-year govt bond rate minus market's inflation forecast)
5. Interest rates — real short-term (3-month) yield
6. Commodities — low price relative to a medium-term average, hoping for mean reversion

Momentum:
1. Stocks — whatever has gone up relatively more in last 11 of the last 12 months, excluding the most recent one (exclusion of the most recent month is pretty standard in the literature for some reasons)
2. Country equity indexes — whatever has gone up more in price in the last 12 months
3. Currencies — ditto
4. Bonds — ditto
5. Interest rates — ditto
6. Commodities — ditto

Carry:
1. Stocks — N/A (high dividend yield kind of captures this but overlaps with value for stocks)
2. Country equity indexes — N/A
3. Currencies — high interest rate on cash onshore
4. Bonds — high rate of 10-year yield relative to cash/T-bill rate, so high roll-down yield
5. Interest rates — high rate of 3-month return (short end of yield curve), so high roll-down yield
6. Commodities — high backwardation (or least contango)

Defensive:
1. Stocks — low beta within market
2. Country equity indexes — low beta compared to other countries (relative to global index)
3. Currencies — N/A
4. Bonds — low beta compared to global index
5. Interest rates — N/A
6. Commodities — N/A

In each of those styles, buy whatever is good by the measure and short whatever looks bad, though in practice this is done systematically and algorithmically across a number of securities. Many of these style/asset combinations will not work a lot of the time. No problem. The hope is to get enough of the exposures and ride them out, with enough of them working enough of the time to turn a profit. The models are only to figure out what should be bought and what sold to represent the style, not how to market time them. And if one wanted, a little more complicated heuristics for any of the above could be used, which I think may be what AQR is actually running in the fund. For example, they and many others propose other measures than simply price/book to get value in stocks.

By the way, the above definitions of value and carry in bonds have a lot of history and should be pretty intuitive, being signals that many talk about around here: you buy when real yields are high and when the yield curve is steep. It is well known that backwardation generally leads to better returns in commodities.

nisiprius wrote:[...]The correlation over that time period was low, only 0.165. However, after reading "skating where the puck was," one has to wonder about persistence. It does not seem to me that it makes sense to apply a fancy strategy that theoretically gives you an 0.20% improvement unless you have already squeezed your weighted-average portfolio expense ratio down into the same neighborhood.[...]

Can anyone think of reasons why in theory these things on a piece-by-piece basis or in aggregate would increase in correlation relative to stocks or bonds over time? This is one of the key questions to ask and consider. For example, has the value factor in stocks gotten more correlated to the stock market (market beta) over time? I don't think I've seen data suggesting anything of the sort.

Now, it could well be that even if styles hold up, transaction costs are more expensive during a crash, as liquidity dries up, and weird things happen.

For one example of what could go wrong, it is very well known that currency carry blows up when there's a shock and equities tank. This has negative skewness with the bad stuff happening at exactly the wrong time, when everyone gets spooked and the exchange rates move against you. This is already seen in the backtest. There may be other such examples that have yet to be realized. However, the construction here is that the impact of any one style/asset combination should be limited, and bets are scaled down as volatility increases, which helps for many crashes as uncertainty usually grows before lightning strikes.

User avatar
Taylor Larimore
Advisory Board
Posts: 25863
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Duplicate

Post by Taylor Larimore » Fri Jun 12, 2015 1:26 pm

Deleted
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
Taylor Larimore
Advisory Board
Posts: 25863
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

A second look at QSPIX

Post by Taylor Larimore » Fri Jun 12, 2015 1:30 pm

Bogleheads:

I have not been following this post since I posted the extra long list of "Risks" on the second page.

I decided to take another look by going to Morningstar, the most reliable source of objective information about mutual funds. This is a small portion of what I found on their website for "AQR Style Premia Alternative I QSPIX.

* QSPIX has a five million dollar investment minimum. (Why are we discussing it?)

* QSPIX has an estimated 10-year cost of $2,944. For comparison, Vanguard Total Stock Market Admiral (VTSAX) has an estimated 10-year cost of $64.

* QSPIX has a 1-year after-tax return of 1.33%. Vanguard VTSAX has a 1-year after-tax return of 10.87%.

* QSPIX is in the bottom 4% of its category year-to-date. Vanguard VTSAX is in the top 35%.

Edit: I forgot to add: Past performance does not guarantee future performance.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
TdF fan
Posts: 119
Joined: Sat Mar 01, 2008 2:35 pm

Re: A second look at QSPIX

Post by TdF fan » Fri Jun 12, 2015 1:53 pm

Taylor Larimore wrote:* QSPIX has a five million dollar investment minimum. (Why are we discussing it?)

Taylor - Fidelity lists a $2,500 minimum for IRA accounts and I think TD Ameritrade has something similar. I'm not advocating for this fund, just pointing out it's more accessible to many than it first appears.

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 12, 2015 1:58 pm

Taylor Larimore wrote:Bogleheads:

I have not been following this post since I posted the extra long list of "Risks" on the second page.

I decided to take another look by going to Morningstar, the most reliable source of objective information about mutual funds. This is a small portion of what I found on their website for "AQR Style Premia Alternative I QSPIX.

* QSPIX has a five million dollar investment minimum. (Why are we discussing it?)

* QSPIX has an estimated 10-year cost of $2,944. For comparison, Vanguard Total Stock Market Admiral (VTSAX) has an estimated 10-year cost of $64.

* QSPIX has a 1-year after-tax return of 1.33%. Vanguard VTSAX has a 1-year after-tax return of 10.87%.

* QSPIX is in the bottom 4% of its category year-to-date. Vanguard VTSAX is in the top 35%.

Best wishes.
Taylor

Many of the concepts are relevant whether or not somebody invests in it. A broader understanding of these factors and underlying returns sources will keep allow skeptics to stay the course and articulate weaknesses of alternative investments from a variety of angles that go deeper than simple cost analysis and hand-selected quotes by experts (which frequently miss the opposite things being said by other experts). A rebalanced three-fund portfolio draws some theoretical underpinnings from modern portfolio theory and an analytical framework that would suggest a large allocation to something like this if returns are anything close to as claimed and expected by some. On the flip side, those investing in this or something similar should understand all the theoretical underpinnings carefully so they have a well-informed plan if the fund does not behave as they currently expect.

The minimum investment and after-tax returns quoted are irrelevant. The way posters are accessing this is through an IRA, as noted above, where the minimum seems to be in the range of $2,500. As we all know, tax-inefficient investments (and this takes the cake) should go in tax-advantaged accounts, where tax drag doesn't matter.

You know as well as anyone that returns should not be judged primarily based on short-term results. It is almost always too early to say. Aside from the returns reported on Morningstar, there is a limited amount of data from before the fund went public in mutual fund format. There are also simulated results by the fund managers for something very close to the actual implementation, which is published in a peer-reviewed journal, with all the strengths and weaknesses of the academic process. There also exist many decades of data for some parts of the strategies. As we all know, this may or may not persist. As with any investment, suitability within the context of a whole portfolio allocation needs to be analyzed, with costs of all kinds necessarily scrutinized, with usually the lowest-cost exposure to the same underlying sources of return preferred.

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

Re: QSPIX - thoughts on interesting fund

Post by countmein » Fri Jun 12, 2015 2:04 pm

Brogleski wrote:
matjen wrote: Or, more relevant to this thread, 3.2 QSPIX, Vanguard VT, and Schwab SCHR. Just spitballing here.

Not trying to pick on this quote, but isn't holding VT and QSPIX at the same time inefficient (from a cost perspective)? It seems to me that one would be able to use SCHR+multi-factor equity funds (PRF,PXF,PXH, etc.) to reach the same loadings as a VT+QSPIX+SCHR portfolio with lower fees.


VT + QSPIX is slightly inefficient because you will be long and short a few of the same names, but this may be so minor as to be inconsequential.

But you cannot replicate VT + QSPIX with RAFI funds because you're only getting one factor with RAFI-- no MOM, defensive, carry, no short side, and no asset diversification.

But setting aside all of QSPIX's extras, from my own back of the envelope cost per total unit of factor exposure analysis, QSPIX and QMNIX 's costs are in line with the cost of DFA core funds, e.g. DFTCX and DFTWX.

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

Re: A second look at QSPIX

Post by countmein » Fri Jun 12, 2015 2:20 pm

Taylor Larimore wrote:Bogleheads:
* QSPIX has a five million dollar investment minimum. (Why are we discussing it?)

the $5M is waived at certain custodians, so it is accessible

* QSPIX has an estimated 10-year cost of $2,944. For comparison, Vanguard Total Stock Market Admiral (VTSAX) has an estimated 10-year cost of $64.

point taken

* QSPIX has a 1-year after-tax return of 1.33%. Vanguard VTSAX has a 1-year after-tax return of 10.87%.

- nobody's holding this in taxable
- QSPIX's annualized return is 8%


* QSPIX is in the bottom 4% of its category year-to-date. Vanguard VTSAX is in the top 35%.

Why compare to VTSAX? That's like comparing BND to VTSAX

Best wishes.
Taylor

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: QSPIX - thoughts on interesting fund

Post by Brogleski » Fri Jun 12, 2015 3:44 pm

Thanks for the response, countmein.

countmein wrote:VT + QSPIX is slightly inefficient because you will be long and short a few of the same names, but this may be so minor as to be inconsequential.
Right, I wasn't too worried about that.

countmein wrote:But you cannot replicate VT + QSPIX with RAFI funds because you're only getting one factor with RAFI-- no MOM, defensive, carry, no short side, and no asset diversification.

I really meant to list the RAFI funds as examples. Couldn't other funds be used to capture those factors in equities (perhaps even other AQR funds)? On the surface, it seems the only truly unique characteristics of QPSIX are ~0 beta, and the asset diversification you alluded to. It just seems like you'd be paying too much for those two characteristics if you're going to add the beta right back in with VT. Maybe I'm misunderstanding something.
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 12, 2015 4:12 pm

Brogleski wrote:I really meant to list the RAFI funds as examples. Couldn't other funds be used to capture those factors in equities (perhaps even other AQR funds)? On the surface, it seems the only truly unique characteristics of QPSIX are ~0 beta, and the asset diversification you alluded to. It just seems like you'd be paying too much for those two characteristics if you're going to add the beta right back in with VT. Maybe I'm misunderstanding something.

Where are you going to get the short side of value, momentum, and low beta from? The long side may be much the same, but not necessarily always.

Also, if you want these at all, why would you want them in only equities and not other classes? And how about carry? If you suppose these factors help, having them together in a fund with 0 beta makes for good rebalancing opportunities.

But what you're asking for on the long side is more or less in iShares's new multifactor ETF line, which has very low AUM right now but does address these things (expense ratio of 0.35% for US large caps):
http://www.ishares.com/us/products/prod ... 2=overview

AQR themselves has them in hedged format in the equity market neutral fund. And long-only in the multi-style funds (Class I in US large caps has an expense ratio capped at 0.45%). Other providers also offer similar products.

Brogleski
Posts: 37
Joined: Fri Jul 12, 2013 4:23 pm

Re: QSPIX - thoughts on interesting fund

Post by Brogleski » Fri Jun 12, 2015 5:03 pm

lack_ey wrote:Where are you going to get the short side of value, momentum, and low beta from? The long side may be much the same, but not necessarily always.
I thought the whole point of the short side of value/momentum was to achieve low beta. Is there another advantage?

lack_ey wrote:Also, if you want these at all, why would you want them in only equities and not other classes? And how about carry? If you suppose these factors help, having them together in a fund with 0 beta makes for good rebalancing opportunities.
Point taken about carry. I never said anything about not wanting asset diversification and low beta. I'm just trying to understand why it would be worth paying so much for them if you're just going to dilute the low beta with pure beta (VT). Just to clarify, I can get behind QSPIX + other tilted funds. I'm just asking about the value(forgive the pun) in QSPIX + VT.

lack_ey wrote:But what you're asking for on the long side is more or less in iShares's new multifactor ETF line, which has very low AUM right now but does address these things (expense ratio of 0.35% for US large caps):
http://www.ishares.com/us/products/prod ... 2=overview

Yeah, I've been keeping my eye on them.
“A good plan, violently executed now, is better than a perfect plan next week.” - George S. Patton

lack_ey
Posts: 5556
Joined: Wed Nov 19, 2014 11:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey » Fri Jun 12, 2015 6:21 pm

Brogleski wrote:
lack_ey wrote:Where are you going to get the short side of value, momentum, and low beta from? The long side may be much the same, but not necessarily always.
I thought the whole point of the short side of value/momentum was to achieve low beta. Is there another advantage?

These "factors" are just regressions based on some certain parameters across the universe of securities. It's just a crude and incomplete decomposition. The momentum factor is defined as the performance of securities that have done relatively well minus the performance of those that have done poorly. Let's say you have a long-only fund that invests in stocks with high momentum and momentum does well in stocks for a year, say because low momentum stocks do terribly while high momentum stocks do below-average and medium momentum does above average. Your fund, which is long only in the high momentum, might do worse than average because the so-called momentum effect being positive was being driven primarily by stocks with low momentum crashing and burning. Sure, a long-only momentum fund would be underweighted in low momentum stocks, but if you were short those stocks, you would have done even better and captured the whole of the momentum effect.

This isn't necessarily a contrived example because at least with value and defensive, a lot of times it is claimed that much of the anomalous performance comes from high growth and high beta doing poorly, rather than value and low beta actually doing better than stocks in the middle.

In any case, to get purer factor exposure as defined academically, you need to be long/short so as to implement the "minus" part of _____ minus _____. It also usually gets greater exposure per dollar invested.

Brogleski wrote:
lack_ey wrote:Also, if you want these at all, why would you want them in only equities and not other classes? And how about carry? If you suppose these factors help, having them together in a fund with 0 beta makes for good rebalancing opportunities.
Point taken about carry. I never said anything about not wanting asset diversification and low beta. I'm just trying to understand why it would be worth paying so much for them if you're just going to dilute the low beta with pure beta (VT). Just to clarify, I can get behind QSPIX + other tilted funds. I'm just asking about the value(forgive the pun) in QSPIX + VT.

Pure beta funds are cheaper and have less turnover, which is good for taxable accounts. You don't run into as much issues with fund providers changing indexes and altering the exposures on you. This sucks if you're stuck with unrealized capital gains in funds you no longer want. I think some small/value tilters may have some horror stories for you. In ETF land, they're a lot more liquid too and trade with tighter spread.

Depending on how much other factor exposure you want relative to market beta, it may or may not be feasible with long-only funds. This was more a problem when multifactor funds were less popular.

Some people may just not want all or a lot of their equities tied to tilting.

Billavoider
Posts: 25
Joined: Sun Jul 07, 2013 6:55 pm

Re: QSPIX - thoughts on interesting fund

Post by Billavoider » Fri Jun 12, 2015 7:32 pm

Incidentally, some of the AQR "I" class funds are available in taxable and non-taxable accounts at Scottrade for a minimum investment of $100.00 plus trade fee.

Post Reply