Author Topic: Dual Momentum Investing  (Read 1940232 times)

Mr. McGibblets

  • 5 O'Clock Shadow
  • *
  • Posts: 64
  • Age: 34
  • Location: Arlington, VA
Dual Momentum Investing
« on: April 14, 2015, 03:55:15 PM »
Hey there -

I read an interesting comment on another thread about Dual Momentum Investing. I have heard of simple momentum investing, but have never heard of investing using the combination of relative strength and absolute momentum.

Here is a great PDF on the topic: http://alpharotation.com/resources/Momentum2%20White%20Paper.pdf

Does anyone have experience with this method? Or concerns/pitfalls?

Cheddar Stacker

  • Magnum Stache
  • ******
  • Posts: 3700
  • Age: 45
  • Location: USA
Re: Dual Momentum Investing
« Reply #1 on: April 14, 2015, 04:43:15 PM »
No experience. I read some of the discussion in the other thread, so I'm an interested follower. Proceed in learning, but with caution would be my advice, just like what was given previously.

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #2 on: April 14, 2015, 06:36:43 PM »
More importantly than understanding how a strategy works, is understanding WHY it works. Unless you have full confidence in a strategy and understand all the ups and downs, you won't stick with it when it's doing poorly. Unseen in the research paper is how many losing trades in a row it experienced. Since it rebalances monthly, it may have 4 or 5 consecutive losing trades. Most people give up after 2 or 3 losing trades, only to see the strategy start working again without them participating.

I recommend reading all you can about it including doing your own testing. Look through the back test on a trade by trade basis and ask yourself, 'is this what I would expect to happen? Would I still trade through this or would I give up?' You need to know it at that level of detail IMO.

FWIW I use my own flavor of dual momentum investing in my retirement and taxable accounts.

The Beacon

  • Stubble
  • **
  • Posts: 148
    • Financial Freedom Tips
Re: Dual Momentum Investing
« Reply #3 on: April 14, 2015, 09:29:02 PM »
More importantly than understanding how a strategy works, is understanding WHY it works. Unless you have full confidence in a strategy and understand all the ups and downs, you won't stick with it when it's doing poorly. Unseen in the research paper is how many losing trades in a row it experienced. Since it rebalances monthly, it may have 4 or 5 consecutive losing trades. Most people give up after 2 or 3 losing trades, only to see the strategy start working again without them participating.

Very true. one time, I suffered 10 consecutive losing trades over the course of 8 days. The psychological impact was unbearable at one point even if I knew my strategy was sound.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Dual Momentum Investing
« Reply #4 on: April 14, 2015, 09:48:08 PM »
Following, but I may hurt my face from palming it so hard if comments like "10 trades in 8 days" are the norm.  ;)
I am a former teacher who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and am now settled with three kids.
If you want to know more about me, this Business Insider profile tells the story pretty well.
I (rarely) blog at AdventuringAlong.com. Check out the Now page to see what I'm up to currently.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #5 on: April 14, 2015, 10:55:34 PM »
I have written an awful lot about dual momentum, because I've thought a lot about dual momentum.

I currently manage all of my tax sheltered  accounts in this manner, and have done so for about 6 months now. For the year before that I managed my HSA account using a form of Dual momentum.

I will include links to all of my articles on the subject below, rather then restating my thoughts sloppily here, but the real expert on the subject is Gary Antonacci Who wrote this book (which I highly recommend.)

http://www.amazon.com/Dual-Momentum-Investing-Innovative-Strategy/dp/0071849440

But before investing in the book I would recommend checking out his blog which is excellent…

http://www.dualmomentum.net/?m=1

Here then are my thoughts on the subject.

On why I personally chose to switch to a dual momentum strategy.

http://www.milesdividendmd.com/jumping-off-a-cliff/

On how dual momentum works...

http://www.milesdividendmd.com/two-faced-investing/

On the barriers to implementing a dual momentum strategy, and paradoxically why I believe dual momentum will persist as a viable strategy in the future.

http://www.milesdividendMD.com/dual-doubts/

And finally on the real world risks of dual momentum as I see them.

http://www.milesdividendmd.com/looking-under-rocks/

Enjoy!
« Last Edit: April 14, 2015, 11:55:27 PM by milesdividendmd »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #6 on: April 14, 2015, 11:35:00 PM »
Following, but I may hurt my face from palming it so hard if comments like "10 trades in 8 days" are the norm.  ;)

That's funny. But that trading frequency does NOT describe DM investing at all!

(I have been using global equities momentum (the simplest form of dual momentum ) in my retirement accounts since last September and have yet to make my first trade.)
« Last Edit: April 14, 2015, 11:57:26 PM by milesdividendmd »

Crushtheturtle

  • 5 O'Clock Shadow
  • *
  • Posts: 51
Re: Dual Momentum Investing
« Reply #7 on: April 15, 2015, 03:57:52 AM »
Milesdividend,

I was wondering if you were the author of those blog posts- very good work. I was referred to your work from a Bogleheads thread. You lay out a very compelling case. In fact, I am executing your exact plan with my Vanguard Roths: S&P500 (VFIAX), Developed Markts (VTMGX), Emerging (VEMAX) and short treasury (VFISX), using a 6 month look back. I am following suit with my TSP funds: G,F,C, and I. (I am excluding the S fund with its medium and small US caps due to its volatility and redundancy to the US market in the C fund).

Question: I presume you chose US, developed International, and Emerging for their geographic diversity? I have unfortunately not read Gary Antonacci's book.


For others, if you google Adaptive or Tactical Asset Allocation, you will find academic papers published over the past few years that pretty much lay out other good arguments for this momentum strategy. Like Miles, I am executing the strategy in only tax sheltered accounts for now, but I'm in a low enough tax bracket that I may go "whole hog" with my taxable acct as well, if this strategy proves as valuable as the evidence suggests.
« Last Edit: April 15, 2015, 03:59:30 AM by Crushtheturtle »

Mr. McGibblets

  • 5 O'Clock Shadow
  • *
  • Posts: 64
  • Age: 34
  • Location: Arlington, VA
Re: Dual Momentum Investing
« Reply #8 on: April 15, 2015, 08:32:34 AM »
Thanks everyone. I just put a hold on the book from my library. I am looking forward to getting into it.

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #9 on: April 15, 2015, 09:34:09 AM »
You can also read Antonacci's research papers here for free: http://www.optimalmomentum.com/papers1.html

I've read through them and have been messing around with backtest spreadsheets I developed independently using historical data from yahoo! finance and am becoming more and more convinced.   I copied my spreadsheet where I implement the strategy described in milesdividendmd's two faced investing post with VFINX, DFALX, DFEMX, and VFISX.  I was stunned that such a simple strategy could avoid the big market downturns in the early 2000s and 2008/2009.

Looking at the spreadsheet (with a 6 month lookback), it puts you in bonds on 1/2008 and puts you back in a stock fund on 1/2009.  You avoid a 33% drawdown during that period in the S&P500.  It has similar success in the dot com bust.

This strategy is really about absolute momentum (read that paper above) and getting completely out of stocks when they aren't doing well.  That of course is market timing which I am generally very wary of.  Usually people are burned badly by buying and selling at exactly the wrong times.  This rule based strategy seems to be an exception to the rule against market timing.  It just plain works.

I think it works on behavioral effects which is why it's so hard to understand logically.  So far I haven't found any fatal flaws and am starting to consider implementing this myself, but more digging needs to be done.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Dual Momentum Investing
« Reply #10 on: April 15, 2015, 09:41:25 AM »
Following, but I may hurt my face from palming it so hard if comments like "10 trades in 8 days" are the norm.  ;)

That's funny. But that trading frequency does NOT describe DM investing at all!

(I have been using global equities momentum (the simplest form of dual momentum ) in my retirement accounts since last September and have yet to make my first trade.)

How frequently have you had to check if a trade was necessary?
I am a former teacher who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and am now settled with three kids.
If you want to know more about me, this Business Insider profile tells the story pretty well.
I (rarely) blog at AdventuringAlong.com. Check out the Now page to see what I'm up to currently.

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #11 on: April 15, 2015, 09:47:37 AM »
The strategy has you check monthly to see if you should make any changes.  Not bad at all.  In the papers I think he says the backtesting averages less than 2 trades per year.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #12 on: April 15, 2015, 09:59:58 AM »

Following, but I may hurt my face from palming it so hard if comments like "10 trades in 8 days" are the norm.  ;)

That's funny. But that trading frequency does NOT describe DM investing at all!

(I have been using global equities momentum (the simplest form of dual momentum ) in my retirement accounts since last September and have yet to make my first trade.)

How frequently have you had to check if a trade was necessary?

Every month. I enjoy it   YMMV!

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #13 on: April 15, 2015, 10:25:57 AM »
Spreadsheet showing the probability of consecutive losses based on winning percentage:
https://docs.google.com/spreadsheets/d/1pHYNz6RV-nx-9Ycz4yoVItLlK0Ur2VdHYTLuW4dzyEc/edit?usp=sharing

FYI most momentum strategies have a winning percentage of 60-80%. The white paper that the OP originally posted has a winning % of ~60-65% based on my own backtests.

FWIW here's my own backtest of what the white paper did. They did not say what their lookback period was (3 months? 6 months? 12 months?) nor did they say if they also screened for volatility or correlation. I kept it simple and used a 6 month lookback for the momentum screen. https://drive.google.com/file/d/0BzyyTlvGE-T2cTFfTzNLSVVjS00/view?usp=sharing

It is rather crude but you can do your own backtest for free using https://www.portfoliovisualizer.com/test-market-timing-model

Here's a spreadsheet you can use to make your own backtest using data from Yahoo. It's a pretty manual spreadsheet but you'll get the idea http://systemtradersuccess.com/backtesting-etf-rotational-system/ and http://systemtradersuccess.com/improving-simple-etf-rotational-model/

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #14 on: April 15, 2015, 10:38:16 AM »
The strategy has you check monthly to see if you should make any changes.  Not bad at all.  In the papers I think he says the backtesting averages less than 2 trades per year.

This is only from 2003 (as far back as ETFReplay will let you go) but this is Gary's implementation of Dual Momentum. Looks like an average of 2.5 trades a year https://drive.google.com/file/d/0BzyyTlvGE-T2Y19FdV9wWXlUaVU/view?usp=sharing

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #15 on: April 15, 2015, 10:46:16 AM »
miles is there a spreadsheet formula you are following to get the asset allocation?

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #16 on: April 15, 2015, 10:58:45 AM »
The strategy has you check monthly to see if you should make any changes.  Not bad at all.  In the papers I think he says the backtesting averages less than 2 trades per year.

This is only from 2003 (as far back as ETFReplay will let you go) but this is Gary's implementation of Dual Momentum. Looks like an average of 2.5 trades a year https://drive.google.com/file/d/0BzyyTlvGE-T2Y19FdV9wWXlUaVU/view?usp=sharing

More fantastic results.  +2% CAGR and half the max drawdown with a very simple strategy. 

I'd love to see what would happen if you implement this in cFIREsim as missing the big drawdowns could significantly mitigate sequence of returns risk.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #17 on: April 15, 2015, 11:09:13 AM »
wait so couple questions...

1. does this dump whole portfolios into one asset class.

sir doug's spreadsheet.  I dont look at ratio's i just dump all of my money one direction.  then enter the closing values at the beginning of each month for the 4 asset classes and see what it spits out?  then proceed to dump all monies that way?


FIPurpose

  • Handlebar Stache
  • *****
  • Posts: 2059
  • Location: ME
    • FI With Purpose
Re: Dual Momentum Investing
« Reply #18 on: April 15, 2015, 11:29:14 AM »
MD,

I have been eating up all this information like crazy. And honestly, I am almost completely convinced. I'm currently looking for reasons not to implement it myself. And like many others I may split my portfolio to where part implements this method and the other part implements a typical buy & hold. I'll be seeing if my library has Gary's book.

Do you currently use the 4 Vanguard funds in your strategy, and does it matter what the 4 funds are? I'm gathering that you need to select between 2-3 equity classes and a bond class of some sort.

More reading is to be done!

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #19 on: April 15, 2015, 11:32:24 AM »
so is all your money in Large cap international right now?

Crushtheturtle

  • 5 O'Clock Shadow
  • *
  • Posts: 51
Re: Dual Momentum Investing
« Reply #20 on: April 15, 2015, 11:49:03 AM »
so is all your money in Large cap international right now?

When I compare my "universe" of 4 funds in a website such as Stockcharts with a look back period of 6 months (meaning adding up the returns of each fund over the previous 126 sessions), the S&P500 is still outperforming Developed Int'l, but just barely. And both are markedly outperforming short term bonds.

So per the strategy, that portion of my portfolio with which I am implementing a momentum strategy will stay concentrated in the S&P until I check again next month.
« Last Edit: April 15, 2015, 11:54:11 AM by Crushtheturtle »

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #21 on: April 15, 2015, 11:50:06 AM »
also with 2 trades a year i guess this isnt a good play for taxable accounts?  realizing those short term gains each year?

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #22 on: April 15, 2015, 11:52:04 AM »
In a straight comparison of VFINX to DFALX i see .28% return for the S&P and .56% return for the Large cap international

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #23 on: April 15, 2015, 12:01:38 PM »

miles is there a spreadsheet formula you are following to get the asset allocation?

No spreadsheet necessary. every month I do a 6 month backtest of total returns on VIIIX, Vea, and shy and allocate 100% to the winner in my 403 B.

I use PERFCHaRTS on stockcharts.com for the backtest.

Crushtheturtle

  • 5 O'Clock Shadow
  • *
  • Posts: 51
Re: Dual Momentum Investing
« Reply #24 on: April 15, 2015, 12:09:34 PM »
In a straight comparison of VFINX to DFALX i see .28% return for the S&P and .56% return for the Large cap international

This is what I'm using:

http://stockcharts.com/freecharts/perf.php?VFIAX,vtmgx,vemax,vfisx

I slide the date range to approx 126 (6 months, 21 sessions per month on average). Results vary with the funds/etfs chosen, as their compositions differ.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #25 on: April 15, 2015, 12:16:47 PM »
Relative momentum is simply an empirical observation that stocks which have recently done will continue to do well in the short term and vice versa.

It's not a risk story, It's not a valuation story. It's probably a behavioral story.

It's Strength is simply the empirical observation that it exists virtually everywhere, in every asset class, and in every market.

Used alone it absolutely improves returns, but offers no protection against drawdowns.

Absolute momentum on the other hand provides a rational signal when to exit the market alltogether, similar to the 200 day moving average with slightly less trading. So it derisks your portfolio with what I term "temporal diversification."  Ie instead of having some bonds all of the time,  you have all bonds part of the time. It is market timing pure and simple, there's no denying this.
« Last Edit: April 15, 2015, 12:19:17 PM by milesdividendmd »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #26 on: April 15, 2015, 12:17:45 PM »

In a straight comparison of VFINX to DFALX i see .28% return for the S&P and .56% return for the Large cap international

This is what I'm using:

http://stockcharts.com/freecharts/perf.php?VFIAX,vtmgx,vemax,vfisx

I slide the date range to approx 126 (6 months, 21 sessions per month on average). Results vary with the funds/etfs chosen, as their compositions differ.

Slide your end date to the end of last month.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #27 on: April 15, 2015, 12:19:18 PM »
gotcha thats a great link turtle  thank you... i can do the same with my 401k account options.  How do fees play a role in all this.  my emerging markets in my 401k blows to the tune of 1.2% fees and the developed foreign  isnt much better at .64% and finally i dont have a SHY eqivalent option include : US Debt index T fund(blackrock .07%) Dodge and cox income fund(.43%)

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #28 on: April 15, 2015, 12:20:07 PM »
Tfund should be fine, skip EM.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #29 on: April 15, 2015, 12:21:44 PM »

In a straight comparison of VFINX to DFALX i see .28% return for the S&P and .56% return for the Large cap international

This is what I'm using:

http://stockcharts.com/freecharts/perf.php?VFIAX,vtmgx,vemax,vfisx

I slide the date range to approx 126 (6 months, 21 sessions per month on average). Results vary with the funds/etfs chosen, as their compositions differ.

Slide your end date to the end of last month.

i still get that VFIAX the S&P is the best place to be ... so is this chart all one really needs to implement this startegy?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #30 on: April 15, 2015, 12:22:57 PM »
Correct. I am now 100% s and p.

Next month may be my first trade. We'll see.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #31 on: April 15, 2015, 12:28:34 PM »

MD,

I have been eating up all this information like crazy. And honestly, I am almost completely convinced. I'm currently looking for reasons not to implement it myself. And like many others I may split my portfolio to where part implements this method and the other part implements a typical buy & hold. I'll be seeing if my library has Gary's book.

Do you currently use the 4 Vanguard funds in your strategy, and does it matter what the 4 funds are? I'm gathering that you need to select between 2-3 equity classes and a bond class of some sort.

More reading is to be done!

I just use VIIIX (S&P), FSPNX (EAFE), and vanguard short term treasuries. I have no suitable EM option in my 403b.

This is what Antonacci terms "global equities momentum," though he uses a total bond fund in place of short term treasuries.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #32 on: April 15, 2015, 12:47:21 PM »
alright i plan to start this next month.  at what expense ratio does this become dumb? my lowest developed is .64% and my bond fund looks like it will be .43 i'll have to look into the other to see if its workable. 

ChaseJuggler

  • Stubble
  • **
  • Posts: 123
Re: Dual Momentum Investing
« Reply #33 on: April 15, 2015, 01:36:56 PM »

FIPurpose

  • Handlebar Stache
  • *****
  • Posts: 2059
  • Location: ME
    • FI With Purpose
Re: Dual Momentum Investing
« Reply #34 on: April 15, 2015, 02:42:30 PM »
So I've been considering what sort of markets are good and bad for buy&hold vs. momentum

So as discussed in your article momentum investing is a poor investment in extremely volatile markets. where every 6 months you are picking the wrong trend.

Though in buy&hold you are betting that no matter what happens, the general trend will be up in the end. Now I have heard several people on this forum bring up the Japanese or even Australian markets as examples of countries that don't move anywhere for several years at a time. Perhaps this is the risk you take on in a buy&hold strategy.

I would be curious in how people analyze and compare those risks.
« Last Edit: April 15, 2015, 03:32:55 PM by flyingcircle »

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #35 on: April 15, 2015, 03:31:13 PM »
I would expect returns of momentum strategies to be similar in the future as they have been in the past. In the past it went through good periods followed by bad periods. The bad periods shake out the people who aren't true believers, and allow the good periods to come back. As long as human nature stays the same, then momentum should be persistent over the long term just as it has in the past.

As you say, buy and hold has it's own risk so in the end, it's what kind of risk are you willing to take.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #36 on: April 15, 2015, 04:16:43 PM »
I would expect returns of momentum strategies to be similar in the future as they have been in the past. In the past it went through good periods followed by bad periods. The bad periods shake out the people who aren't true believers, and allow the good periods to come back. As long as human nature stays the same, then momentum should be persistent over the long term just as it has in the past.

As you say, buy and hold has it's own risk so in the end, it's what kind of risk are you willing to take.

While this observation of the popularity of an approach crowding out the future performance is true of other investment approaches like the value approachs or the dividend approach,    I am not sure that it is true of the momentum approach. After all if more people crowd into momentum, what are you left with?

You are left with more upwards price movement, a.k.a. more momentum. 

There is certainly the risk of a bubble forming, but as long as there is sufficient liquidity in short-term treasuries to trade out of equities, absolute momentum should protect against the downside risk of the bubble bursting better than for buy and holders.
« Last Edit: April 15, 2015, 04:21:29 PM by milesdividendmd »

FIPurpose

  • Handlebar Stache
  • *****
  • Posts: 2059
  • Location: ME
    • FI With Purpose
Re: Dual Momentum Investing
« Reply #37 on: April 15, 2015, 04:47:43 PM »
I'll be going through a number of steps this weekend and have a convo with the spouse, but I think I've read enough to know the risks and trade offs. My portfolio is still decades from draw down, so I believe this strategy to be a winner long term over buy & hold. If people are interested I'll start a journal exploring how this method compares to how a buy & hold strategy would compare.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #38 on: April 15, 2015, 05:02:48 PM »

I'll be going through a number of steps this weekend and have a convo with the spouse, but I think I've read enough to know the risks and trade offs. My portfolio is still decades from draw down, so I believe this strategy to be a winner long term over buy & hold. If people are interested I'll start a journal exploring how this method compares to how a buy & hold strategy would compare.

There is little to be lost (and much to be gained) by transitioning over in stages.

Maybe splitting your tax sheltered portfolio into 2 buckets?  Buy and hold Vs DM.

When all of your wealth is in a single index, the volatility can take some getting used to. ( I invested my HSA account only in this way for a year before switching over which, in retrospect, was helpful. )

691175002

  • 5 O'Clock Shadow
  • *
  • Posts: 69
Re: Dual Momentum Investing
« Reply #39 on: April 15, 2015, 05:04:29 PM »
As with many investing strategies, knowledge generally leads to further uncertainty.

The dual momentum blog and book provide a fairly one-sided analysis of momentum and many of the back-tests are misrepresented as they incorporate varying amounts of future information.

Momentum is one of the original factors and has been extensively studied.  It is well understood to be the strongest market anomaly.

http://www.msci.com/resources/pdfs/Foundations_of_Factor_Investing.pdf

Significant analysis has also been done on momentum crashes.  Momentum is one of the few investment strategies that creates positive feedback (where investors piling on acts as a signal for more investors to enter the trade).  This ends up creating infrequent but brutal drawdowns, which is one of the reasons momentum investing is considered a risk premium and not free money.

There are people who claim you can avoid momentum crashes by overlaying a second signal or diversifying in other ways but those systems are almost always fit to historical data in-sample and there is no expectation they should be robust in the future.  Standard deviation is not an effective way of quantifying a skewed return profile.

I am a believer in momentum investing, but am aware that excess returns are compensation for risk, not a free lunch.  There are reasons to believe momentum will generate superior returns if you can stay disciplined.

Strategies that appear to have abnormally consistent returns (writing options, momentum, statistical arbitrage) generally conceal infrequent but large losses.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #40 on: April 15, 2015, 05:39:54 PM »
so is DMSR the rotation of the 4 asset classes that includes emerging markets where as GEMs is just the 3 not including emerging markets.  Does anyone have a DMSR model?

Financial.Velociraptor

  • Handlebar Stache
  • *****
  • Posts: 2144
  • Age: 51
  • Location: Houston TX
  • Devour your prey raptors!
    • Living Universe Foundation
Re: Dual Momentum Investing
« Reply #41 on: April 15, 2015, 06:59:10 PM »

Strategies that appear to have abnormally consistent returns (writing options, momentum, statistical arbitrage) generally conceal infrequent but large losses.

This.  I do very well writing options.  But I got burned heavily a few times while I was learning.  Now, I write puts only on companies I already want to own at a price I'd be willing pay for buy/hold.  My "downside" is thus tolerable when something tanks and I end up holding something with short term capital losses.  The result is a blue-chip intensive strategy that focuses on companies that have leading P/E, P/B, or P/S within their industry.  I get burned far less often now, and the skid marks are much shorter than when I chased yield by writing puts on Chinese solar companies with negative earnings but fat premiums.

I'd do back-testing with a mechanical trailing stop loss applied to see if there is a sweet spot that gets you out before market collapses before putting real money on DM.

The Beacon

  • Stubble
  • **
  • Posts: 148
    • Financial Freedom Tips
Re: Dual Momentum Investing
« Reply #42 on: April 15, 2015, 07:17:42 PM »
As with many investing strategies, knowledge generally leads to further uncertainty.

The dual momentum blog and book provide a fairly one-sided analysis of momentum and many of the back-tests are misrepresented as they incorporate varying amounts of future information.

Momentum is one of the original factors and has been extensively studied.  It is well understood to be the strongest market anomaly.

http://www.msci.com/resources/pdfs/Foundations_of_Factor_Investing.pdf

Significant analysis has also been done on momentum crashes.  Momentum is one of the few investment strategies that creates positive feedback (where investors piling on acts as a signal for more investors to enter the trade).  This ends up creating infrequent but brutal drawdowns, which is one of the reasons momentum investing is considered a risk premium and not free money.

There are people who claim you can avoid momentum crashes by overlaying a second signal or diversifying in other ways but those systems are almost always fit to historical data in-sample and there is no expectation they should be robust in the future.  Standard deviation is not an effective way of quantifying a skewed return profile.

I am a believer in momentum investing, but am aware that excess returns are compensation for risk, not a free lunch.  There are reasons to believe momentum will generate superior returns if you can stay disciplined.

Strategies that appear to have abnormally consistent returns (writing options, momentum, statistical arbitrage) generally conceal infrequent but large losses.
Very good analysis.  if you chase returns, then be prepared for the bigger than normal draw-downs. I would not risk my whole portfolio on anything that has abnormal high returns on paper.  To be honest,  my main strategy has an average worst annual draw-down of 4.5% on paper after thousands of trades. I still do not have the balls to put all my money in, 30% at most. I wish I could scale up because the results on paper make me drool........ But I have no problem with 100% in an index fund...

FIPurpose

  • Handlebar Stache
  • *****
  • Posts: 2059
  • Location: ME
    • FI With Purpose
Re: Dual Momentum Investing
« Reply #43 on: April 15, 2015, 08:37:11 PM »

I'll be going through a number of steps this weekend and have a convo with the spouse, but I think I've read enough to know the risks and trade offs. My portfolio is still decades from draw down, so I believe this strategy to be a winner long term over buy & hold. If people are interested I'll start a journal exploring how this method compares to how a buy & hold strategy would compare.

There is little to be lost (and much to be gained) by transitioning over in stages.

Maybe splitting your tax sheltered portfolio into 2 buckets?  Buy and hold Vs DM.

When all of your wealth is in a single index, the volatility can take some getting used to. ( I invested my HSA account only in this way for a year before switching over which, in retrospect, was helpful. )

Oh no doubt,

I'll be using about 1/3 of my portfolio to see how I feel about this strategy, and also see how I handle the swings. Thanks for your articles. They were an immense help.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #44 on: April 15, 2015, 08:45:24 PM »

so is DMSR the rotation of the 4 asset classes that includes emerging markets where as GEMs is just the 3 not including emerging markets.  Does anyone have a DMSR model?

DMSR is dual momentum sector rotation.instead of rotating between asset classes it rotates between sectors within the US market based on dual momentum.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #45 on: April 15, 2015, 09:00:46 PM »

As with many investing strategies, knowledge generally leads to further uncertainty.

The dual momentum blog and book provide a fairly one-sided analysis of momentum and many of the back-tests are misrepresented as they incorporate varying amounts of future information.

Momentum is one of the original factors and has been extensively studied.  It is well understood to be the strongest market anomaly.

http://www.msci.com/resources/pdfs/Foundations_of_Factor_Investing.pdf

Significant analysis has also been done on momentum crashes.  Momentum is one of the few investment strategies that creates positive feedback (where investors piling on acts as a signal for more investors to enter the trade).  This ends up creating infrequent but brutal drawdowns, which is one of the reasons momentum investing is considered a risk premium and not free money.

There are people who claim you can avoid momentum crashes by overlaying a second signal or diversifying in other ways but those systems are almost always fit to historical data in-sample and there is no expectation they should be robust in the future.  Standard deviation is not an effective way of quantifying a skewed return profile.

I am a believer in momentum investing, but am aware that excess returns are compensation for risk, not a free lunch.  There are reasons to believe momentum will generate superior returns if you can stay disciplined.

Strategies that appear to have abnormally consistent returns (writing options, momentum, statistical arbitrage) generally conceal infrequent but large losses.

I don't mean to sound harsh, but I actually don't find this criticism to be very enlightening.

I am very open to specific criticisms of this (or any) strategy.  But this one seems so general as to be useless.

If you talk about momentum as above then what you are talking about is relative momentum or price momentum.

Dual momentum is a very different animal by virtue of absolute momentum.

And absolute momentum is just a very specific type of trend following. It is not so different from the 200 day moving average, and really the only difference is that it trades a little bit less frequently.

For this reason if you combine the 200 day moving average approach  with relative momentum you get very similar results to a dual momentum strategy.

This is because trend following almost universally limits drawdowns. There are really no exceptions to this rule over a long time horizons.

Over short time horizons trend following certainly is susceptible to whipsaws and flash crashes. (Which I write about in my  last article, but The downside of such an approach is limited to the actual drawdown of the flash crash, and  underperformance relative to the market. My fundamental understanding of such an approach is that it is axiomatic that if you can limit your maximum draw down to 20 to 30% without giving up most of the upside of equities you will do incredibly well long-term.

To dismiss this strategy as performance chasing just strikes me as somewhat lazy and generic. I.e. you can make this criticism about any strategy without really thinking about the specific problems of the strategy being discussed.

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #46 on: April 15, 2015, 10:18:35 PM »

As with many investing strategies, knowledge generally leads to further uncertainty.

The dual momentum blog and book provide a fairly one-sided analysis of momentum and many of the back-tests are misrepresented as they incorporate varying amounts of future information.

Momentum is one of the original factors and has been extensively studied.  It is well understood to be the strongest market anomaly.

http://www.msci.com/resources/pdfs/Foundations_of_Factor_Investing.pdf

Significant analysis has also been done on momentum crashes.  Momentum is one of the few investment strategies that creates positive feedback (where investors piling on acts as a signal for more investors to enter the trade).  This ends up creating infrequent but brutal drawdowns, which is one of the reasons momentum investing is considered a risk premium and not free money.

There are people who claim you can avoid momentum crashes by overlaying a second signal or diversifying in other ways but those systems are almost always fit to historical data in-sample and there is no expectation they should be robust in the future.  Standard deviation is not an effective way of quantifying a skewed return profile.

I am a believer in momentum investing, but am aware that excess returns are compensation for risk, not a free lunch.  There are reasons to believe momentum will generate superior returns if you can stay disciplined.

Strategies that appear to have abnormally consistent returns (writing options, momentum, statistical arbitrage) generally conceal infrequent but large losses.

I understand what you are saying but the data you show here is a long/short individual stock momentum system with no trend filter vs a cross-asset momentum system w/ filter. Not exactly apples to apples. Long/short momentum applied to only individual stocks is pretty scary in simple form. The crashes don't come from momentum stocks crashing, it actually comes from the shorted stocks spiking upwards after a market crash. Men Faber showed that by simply lowering the shorted stocks by 20% for every 10% decline in the market, it effectively wiped out the previous momentum crashes from the backtest. Of course, you could also just apply momentum in a long-only basis and skip the crashes as well.

Trend Following with Managed Futures: The Search for Crisis Alpha (Wiley Trading) https://www.amazon.com/dp/1118890973/ref=cm_sw_r_awd_RsZlvb04E7KX8
https://www.amazon.com/dp/1118890973/ref=cm_sw_r_awd_RsZlvb04E7KX8

On top of Gary's book, I highly recommend this one. Dual momentum is really just a variation of diversified trend following which has been used successfully (with the fund track records to prove it) since at least the 1970s. Before that guys were using it but typically at the individual stock or commodity level. Alex does a very simple trend following strategy of using only 'absolute momentum' of 12 months applied to every market they could get data for back to 1200 a.d. If 12 month momentum was positive, they went long, when it was negative, they went short. Each market was equal weighted according to its volatility and long term returns were 13% with a max drawdown of less than 30% over 800 years. Even more interesting is the fact that it had equity like returns but with zero correlation. This allows you to combine it with long only equities and create a portfolio with the same or better returns, and with lower volatility and drawdowns. Returns are also much more consistent.

The Beacon

  • Stubble
  • **
  • Posts: 148
    • Financial Freedom Tips
Re: Dual Momentum Investing
« Reply #47 on: April 15, 2015, 10:48:51 PM »
To dismiss this strategy as performance chasing just strikes me as somewhat lazy and generic. I.e. you can make this criticism about any strategy without really thinking about the specific problems of the strategy being discussed.
Actually, there is nothing wrong with chasing risk adjusted performance. Different people have different appetite for draw downs/risk. It is safe to say that risk/reward is proportional in most cases. For me, if a system has abnormal returns with limited draw downs on paper, the first thought I have is if it is curve fitting.

No matter how pretty a strategy looks on paper, the only way to tell is  to put it to test in real trading. Paper trading is totally different than real trading. Once the psychological force kicks in, things will suddenly look so different than what they used to look on paper even if they are actually the same.  I rode hard this emotional roller coaster in Jan. 2015 after I scaled up my position size 4 times as big as last year. Originally I was going to try 10 times because it looked so beautiful on paper from all different angles. I am glad I did not..

I hope your strategy works for you in real trading. If it does, we can take a page from you. Good Luck.
 


milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #48 on: April 15, 2015, 11:33:01 PM »
To dismiss this strategy as performance chasing just strikes me as somewhat lazy and generic. I.e. you can make this criticism about any strategy without really thinking about the specific problems of the strategy being discussed.
Actually, there is nothing wrong with chasing risk adjusted performance. Different people have different appetite for draw downs/risk. It is safe to say that risk/reward is proportional in most cases. For me, if a system has abnormal returns with limited draw downs on paper, the first thought I have is if it is curve fitting.

No matter how pretty a strategy looks on paper, the only way to tell is  to put it to test in real trading. Paper trading is totally different than real trading. Once the psychological force kicks in, things will suddenly look so different than what they used to look on paper even if they are actually the same.  I rode hard this emotional roller coaster in Jan. 2015 after I scaled up my position size 4 times as big as last year. Originally I was going to try 10 times because it looked so beautiful on paper from all different angles. I am glad I did not..

I hope your strategy works for you in real trading. If it does, we can take a page from you. Good Luck.

Thanks Sharpy. I truly wish you well in your investing too.

I have only converted my retirement accounts to this strategy for six months, so it's impossible to draw any conclusions.  But so far so good.

I have increased my exposure to stocks, made no trades, and both of these have been beneficial to my portfolio thus far.

The thing that initially appealed to me about dual momentum, was the exact same thing that has always made me very leery of  using leverage. I am a coward.  I am risk-averse.I am anti-drawdown.

My basic feeling is that a long-term investing strategy has to first and foremost avoid blowups to be sustainable and successful.

What was most attractive to me about dual momentum is not that it returns more than the market during bull markets (usually), it is that it (almost always) loses much less during bear markets.

I believe that this is what Warren Buffett was getting at when he said that the first rule of investing was: don't lose money and that the second rule of investing was: see rule number one.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Dual Momentum Investing
« Reply #49 on: April 16, 2015, 05:47:07 AM »
You can also read Antonacci's research papers here for free: http://www.optimalmomentum.com/papers1.html

I've read through them and have been messing around with backtest spreadsheets I developed independently using historical data from yahoo! finance and am becoming more and more convinced.   I copied my spreadsheet where I implement the strategy described in milesdividendmd's two faced investing post with VFINX, DFALX, DFEMX, and VFISX.  I was stunned that such a simple strategy could avoid the big market downturns in the early 2000s and 2008/2009.

Looking at the spreadsheet (with a 6 month lookback), it puts you in bonds on 1/2008 and puts you back in a stock fund on 1/2009.  You avoid a 33% drawdown during that period in the S&P500.  It has similar success in the dot com bust.

This strategy is really about absolute momentum (read that paper above) and getting completely out of stocks when they aren't doing well.  That of course is market timing which I am generally very wary of.  Usually people are burned badly by buying and selling at exactly the wrong times.  This rule based strategy seems to be an exception to the rule against market timing.  It just plain works.

I think it works on behavioral effects which is why it's so hard to understand logically.  So far I haven't found any fatal flaws and am starting to consider implementing this myself, but more digging needs to be done.

Just an FYI your spreadsheet is a 5 month lookback thats why i was getting different answers than when using the site that MilesDividend recommended