Author Topic: Dual Momentum Investing  (Read 1938834 times)

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #700 on: August 26, 2015, 07:20:25 PM »


Arguing that the market is inefficient as you have done here Forummm is supportive of the central thesis of momentum investing, and counter to the random walk theory from which buy and hold indexing is originated.

And don't take my word for it. Take Sol's. Here is his quote from 8/22 in another thread.

"Quote from: Pooperman on August 22, 2015, 10:05:22 AM
If markets were rational, there would never be sales or spikes.

I'm not so sure about this.  People always talk about the 2000 tech bubble/crash as an obvious example of irrational markets, but it seemed to make perfect sense at the time.  Prices are based on people's expectations of future profits, so when expectations change prices change.  Expectations can change for all sorts of reasons.

Last month, the prevailing view in the financial pages was that megacorp tech firms were dominating the markets because they are so insanely profitable.  Apple is worth more than Exxon/Mobil but it has 200 billion in cash and a P/E of only 13, so it must still be a good value, right?  Fast forward a month and China devalues their currency and Iran is dumping oil during a glut and the fed is raising interest rates, and suddenly everyone's expectations for the future are not so rosy anymore.

Nothing really changed over that month other than how people think the future will unfold, given some new information.  It seems perfectly rational to me for people one month ago to expect Apple and Google and Amazon and Facebook to continue to mint their own money.  It also seems perfectly rational to me for people today to think profits will mean-revert.  A 10% price swing on those rational expectations doesn't strike me as anything freakishly unusual."

Huh?  Forummm's point and Sol's point (in the quoted text from the other thread) seem to be diametrically opposed.  Forummm's point seemed to be that these wild market gyrations were irrational, having no basis in any underlying change in material information (which, as you pointed out, is inconsistent with the efficient market hypothesis), while Sol's point was that even market gyrations that might appear to be irrational are in fact not so, because they can be explained by the incorporation of new material information into the market's expectations (which is consistent with the EMH).

Exactly Brooklyn. That was the point.

Forummm is inexplicably arguing against DM  by arguing against efficient markets. Everything about this argument is counterproductive.

Sol, IMO, is wrong. But at least he is internally consistent.

Forummm was just being funny. The markets are both rational and irrational. You have lots of people trying to get an edge and guess where things are going. You have lots of people who jump in and out of positions. You have lots of people reacting and overreacting to news. In the long run if you B&H you make the market return, and don't care about all this gaming. If you DM, it's unclear what happens to you, and you are part of this attempt at gaming. Maybe your signals point in the right direction, maybe they lead you astray.

My illustration of the short term (hours) seemingly random fluctuations is not necessarily instructive of a 6 or 12 month long lookback period. However, I think that there is clearly some momentum that does take place from time to time. But being able to capitalize on it using a backtested formula is another question. Especially when that formula lags B&H half the decades.

I've already addressed the internal inconsistency of your "being funny", so i'll ignore that.

Which data are you referring to when you say that Dual momentum lags B&H for half of decades?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #701 on: August 26, 2015, 07:26:16 PM »

For what it's worth, my lookback is 6 months and my day to check is Monday. VTI needs to shoot up about 7% by then, or I'm switching to 100% bonds.

You're halfway there today.  Back up 7% from yesterday seems like a real possibility.  So does down another 5%.

So how many DM traders here have actually switched asset classes?  It sure looks like some lookback periods suggest you should, but I don't have enough faith in continued declines to adhere to such a strategy.  Not when the overall economy looks so strong and the market price declines look like panic selling.

My day is next week. Odds are I go to short term treasuries for the month and I'm comfortable with that.

No faith required. If I thought that my opinion of where the market was going to go had any relevance to my likelihood of investment success, then neither buy and hold, nor DM would be my choice of strategies.  In that case I would choose to be an active stock picker who took advantage of my superior foretelling ability.

forummm

  • Walrus Stache
  • *******
  • Posts: 7374
  • Senior Mustachian
Re: Dual Momentum Investing
« Reply #702 on: August 26, 2015, 07:57:28 PM »


Arguing that the market is inefficient as you have done here Forummm is supportive of the central thesis of momentum investing, and counter to the random walk theory from which buy and hold indexing is originated.

And don't take my word for it. Take Sol's. Here is his quote from 8/22 in another thread.

"Quote from: Pooperman on August 22, 2015, 10:05:22 AM
If markets were rational, there would never be sales or spikes.

I'm not so sure about this.  People always talk about the 2000 tech bubble/crash as an obvious example of irrational markets, but it seemed to make perfect sense at the time.  Prices are based on people's expectations of future profits, so when expectations change prices change.  Expectations can change for all sorts of reasons.

Last month, the prevailing view in the financial pages was that megacorp tech firms were dominating the markets because they are so insanely profitable.  Apple is worth more than Exxon/Mobil but it has 200 billion in cash and a P/E of only 13, so it must still be a good value, right?  Fast forward a month and China devalues their currency and Iran is dumping oil during a glut and the fed is raising interest rates, and suddenly everyone's expectations for the future are not so rosy anymore.

Nothing really changed over that month other than how people think the future will unfold, given some new information.  It seems perfectly rational to me for people one month ago to expect Apple and Google and Amazon and Facebook to continue to mint their own money.  It also seems perfectly rational to me for people today to think profits will mean-revert.  A 10% price swing on those rational expectations doesn't strike me as anything freakishly unusual."

Huh?  Forummm's point and Sol's point (in the quoted text from the other thread) seem to be diametrically opposed.  Forummm's point seemed to be that these wild market gyrations were irrational, having no basis in any underlying change in material information (which, as you pointed out, is inconsistent with the efficient market hypothesis), while Sol's point was that even market gyrations that might appear to be irrational are in fact not so, because they can be explained by the incorporation of new material information into the market's expectations (which is consistent with the EMH).

Exactly Brooklyn. That was the point.

Forummm is inexplicably arguing against DM  by arguing against efficient markets. Everything about this argument is counterproductive.

Sol, IMO, is wrong. But at least he is internally consistent.

Forummm was just being funny. The markets are both rational and irrational. You have lots of people trying to get an edge and guess where things are going. You have lots of people who jump in and out of positions. You have lots of people reacting and overreacting to news. In the long run if you B&H you make the market return, and don't care about all this gaming. If you DM, it's unclear what happens to you, and you are part of this attempt at gaming. Maybe your signals point in the right direction, maybe they lead you astray.

My illustration of the short term (hours) seemingly random fluctuations is not necessarily instructive of a 6 or 12 month long lookback period. However, I think that there is clearly some momentum that does take place from time to time. But being able to capitalize on it using a backtested formula is another question. Especially when that formula lags B&H half the decades.

I've already addressed the internal inconsistency of your "being funny", so i'll ignore that.

Which data are you referring to when you say that Dual momentum lags B&H for half of decades?

Analysis we did several hundred posts ago on this thread using 140 years of Schiller market data.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #703 on: August 26, 2015, 08:04:23 PM »
Post number?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #704 on: August 26, 2015, 10:44:31 PM »
Fellow DMers.

Here's an interesting article for those who feel confident that the market will recover from here.  (my basic assumption.)

This article is neither Bearish Nor bullish. (What's the correct animal for ambivalent? Platypus-ish)

But the Bearish comments of Joel Greenblatt certainly make me feel better about my probable eminent move to short-term treasuries.

http://m.huffpost.com/us/entry/8039312

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #705 on: August 26, 2015, 10:59:37 PM »
The decade by decade analysis was AM not DM.

Thanks for the article. I've been expecting to move out of equities for a while now as they have been flat all year and therefore showing very little positive momentum.  It seems like the market just needed a reason for a sell off and China gave one.  This part of the article basically talks about momentum:

"Buyers always appear ready to step in looking for an opportunity. What makes this time a little different is that the market was showing signs of fatigue. The S&P 500 made its all-time high on May 19, and on several occasions failed to make a new high after rebounding from short-term sell-offs. Also, the Chinese problems are serious and not going away."


Sent from my iPhone using Tapatalk

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #706 on: August 26, 2015, 11:09:59 PM »
Duplicate
« Last Edit: August 26, 2015, 11:20:53 PM by milesdividendmd »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #707 on: August 26, 2015, 11:19:36 PM »

The decade by decade analysis was AM not DM.

Thanks for the article. I've been expecting to move out of equities for a while now as they have been flat all year and therefore showing very little positive momentum.  It seems like the market just needed a reason for a sell off and China gave one.  This part of the article basically talks about momentum:

"Buyers always appear ready to step in looking for an opportunity. What makes this time a little different is that the market was showing signs of fatigue. The S&P 500 made its all-time high on May 19, and on several occasions failed to make a new high after rebounding from short-term sell-offs. Also, the Chinese problems are serious and not going away."


Sent from my iPhone using Tapatalk

Yeah. That makes more sense about the decade by decade analysis, since DM Backtesting does not support this assertion at all at least going back to 1971.

Absolute momentum is all about downside protection so it only outperforms when there are bear markets (but always diminishes drawdowns over long time horizons.) So bullish decades are almost sure to sport mild underperformance.

That being said, since the the over performance in bear markets far eclipses the small loss of upside of in bull markets  even absolute momentum alone is a good deal over long time horizons.  DM is just better.

forummm

  • Walrus Stache
  • *******
  • Posts: 7374
  • Senior Mustachian
Re: Dual Momentum Investing
« Reply #708 on: August 27, 2015, 10:40:50 AM »
The decade by decade analysis was AM not DM.

Thanks for the article. I've been expecting to move out of equities for a while now as they have been flat all year and therefore showing very little positive momentum.  It seems like the market just needed a reason for a sell off and China gave one.  This part of the article basically talks about momentum:

"Buyers always appear ready to step in looking for an opportunity. What makes this time a little different is that the market was showing signs of fatigue. The S&P 500 made its all-time high on May 19, and on several occasions failed to make a new high after rebounding from short-term sell-offs. Also, the Chinese problems are serious and not going away."

So what's the explanation now that the market has gone up about 7% in the past 30 hours? No more fatigue? The momentum was down, and now is it up again?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #709 on: August 27, 2015, 10:58:52 AM »
The momentum is still negative on a 6 month lookback, and not by a little bit.

You seem to be under the opinion that one must have an opinion about why the market moves in the way it does in order to practice DM. You don't, any more than you need such knowledge for buy and hold.

In fact as previously discussed buy and hold is much more rooted in the EMH.  DM is the other side of efficient market hypothesis.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Dual Momentum Investing
« Reply #710 on: August 27, 2015, 11:51:58 AM »
I look forward to reexamining this thread over time. I am very curious to see how a DM approach handles this increased volatility.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Dual Momentum Investing
« Reply #711 on: August 27, 2015, 12:01:02 PM »
I look forward to reexamining this thread over time. I am very curious to see how a DM approach handles this increased volatility.

I think it will be hard to say much because of the details of implementation.  If your decision date was a few days ago and your lookback period was short, you definitely switched to Tbills or cash or something.  If your date is next week and your lookback is short you probably sit tight.  If your lookback is long you probably switch eventually, but the performance result you get after that decision will strongly depend on exactly when you make the call, since volatility is so high right now.

Even if somebody switches and then gets whipsawed, I think a die hard will just swallow hard and accept it as par for the strategy, slight underperformance as the cost of trying to reduce big drawdowns.  Buy and holders do the same thing in a big drop, telling themselves the losses are an expected part of their strategy and that they too just need to stick it out.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #712 on: August 27, 2015, 12:26:54 PM »
This is mostly right Sol.

One small point:  If today were your trading day, your lookback could be anything from 3 to 12 months and you would still move to short term treasuries.  This nicely illustrates why the particular lookback  period is not terribly important.  A significant downturn causes a surprisingly consistent signal regardless of the lookback you choose.

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #713 on: August 27, 2015, 02:10:58 PM »
The decade by decade analysis was AM not DM.

Thanks for the article. I've been expecting to move out of equities for a while now as they have been flat all year and therefore showing very little positive momentum.  It seems like the market just needed a reason for a sell off and China gave one.  This part of the article basically talks about momentum:

"Buyers always appear ready to step in looking for an opportunity. What makes this time a little different is that the market was showing signs of fatigue. The S&P 500 made its all-time high on May 19, and on several occasions failed to make a new high after rebounding from short-term sell-offs. Also, the Chinese problems are serious and not going away."

So what's the explanation now that the market has gone up about 7% in the past 30 hours? No more fatigue? The momentum was down, and now is it up again?

This strategy is not a heavy trading strategy.  Most years have less than a handful of trades.

A 3-12 month lookback period and only monthly changes is key to keeping trading costs down and avoiding whipsawing.  This is not that different than looking at the smooth 200 day moving average compared to the ragged daily or hourly data but you don't look at the average every second of every day, just once per month.

I'm really not interested in intra-day or even intra-week moves.  My time horizon is 10 years and I am trying to avoid multi-month bear markets. 

Through my reading and self-performed backtesting I am convinced of the value and downside protection of this type of rules based dual momentum and will follow the 6 month signals once a month on the third Friday of each month. 

forummm

  • Walrus Stache
  • *******
  • Posts: 7374
  • Senior Mustachian
Re: Dual Momentum Investing
« Reply #714 on: August 27, 2015, 04:58:33 PM »
The decade by decade analysis was AM not DM.

Thanks for the article. I've been expecting to move out of equities for a while now as they have been flat all year and therefore showing very little positive momentum.  It seems like the market just needed a reason for a sell off and China gave one.  This part of the article basically talks about momentum:

"Buyers always appear ready to step in looking for an opportunity. What makes this time a little different is that the market was showing signs of fatigue. The S&P 500 made its all-time high on May 19, and on several occasions failed to make a new high after rebounding from short-term sell-offs. Also, the Chinese problems are serious and not going away."

So what's the explanation now that the market has gone up about 7% in the past 30 hours? No more fatigue? The momentum was down, and now is it up again?

This strategy is not a heavy trading strategy.  Most years have less than a handful of trades.

A 3-12 month lookback period and only monthly changes is key to keeping trading costs down and avoiding whipsawing.  This is not that different than looking at the smooth 200 day moving average compared to the ragged daily or hourly data but you don't look at the average every second of every day, just once per month.

I'm really not interested in intra-day or even intra-week moves.  My time horizon is 10 years and I am trying to avoid multi-month bear markets. 

Through my reading and self-performed backtesting I am convinced of the value and downside protection of this type of rules based dual momentum and will follow the 6 month signals once a month on the third Friday of each month. 

I get that DM is a longer term lookback. I was just asking about the specific explanation you quoted in the article. I find it interesting that humans have a need to sort of anthropomorphise* the market and explain why it did this or that because of its feelings about this or that. But oftentimes their explanations aren't very consistent from hour to hour.

*yeah it's shorthand for the people and machines doing the trades based on whatever feelings, but a funny (to me) practice nonetheless.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Dual Momentum Investing
« Reply #715 on: August 28, 2015, 02:33:27 PM »
This is mostly right Sol.

One small point:  If today were your trading day, your lookback could be anything from 3 to 12 months and you would still move to short term treasuries.  This nicely illustrates why the particular lookback  period is not terribly important.  A significant downturn causes a surprisingly consistent signal regardless of the lookback you choose.
I disagree. I examined the performance of DM over the past 25 years, using a tool posted early in the thread, and your lookback period had a massive effect. The difference between a 3, 6 and 12 month lookback was considerable (with 3 month being the best performer, and 12 month the worst).

This difference, and the apparent arbitrary nature of it, is the reason that I initially chose not to employ this strategy.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #716 on: August 28, 2015, 02:51:10 PM »

This is mostly right Sol.

One small point:  If today were your trading day, your lookback could be anything from 3 to 12 months and you would still move to short term treasuries.  This nicely illustrates why the particular lookback  period is not terribly important.  A significant downturn causes a surprisingly consistent signal regardless of the lookback you choose.
I disagree. I examined the performance of DM over the past 25 years, using a tool posted early in the thread, and your lookback period had a massive effect. The difference between a 3, 6 and 12 month lookback was considerable (with 3 month being the best performer, and 12 month the worst).

This difference, and the apparent arbitrary nature of it, is the reason that I initially chose not to employ this strategy.

Interesting claim. Please share your data.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Dual Momentum Investing
« Reply #717 on: August 28, 2015, 03:14:00 PM »
Interesting claim. Please share your data.

He said he only did 25 years.  I'm sure the true believers can safely ignore his results as insufficiently robust, if they disagree with the conclusions.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #718 on: August 28, 2015, 03:24:51 PM »
I'd prefer to see the data before commenting.

Did all lookbacks outperform B and H?, were transaction costs included? Were drawdowns similar? Were drawdowns all superior to B and H?

The description in the post in question tells me nothing but that Chuck thought the results were different.

But I do agree 25 years is a short time period for Backtesting.

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #719 on: August 28, 2015, 03:53:21 PM »
Hi, I'm a European investor, and I'm considering this strategy for part of my invested money.

Does anyone have recommendations for international investors? Is there anyone already using this strategy?
I'm specifically looking for info on which funds, ETFs or indices I should use, and which lookback periods seem appropriate. Any ideas? Since US and European markets are pretty correlated, does it make sense to e.g. choose S&P500 and EuroStoxx600 (plus some emerging markets for example)?
Due to currency valuations there was quite a different performance this year, so maybe the correlation isn't that big after all if one considers the currency exchange ratios.

I know Mr. Antonacci has a post on his blog about how to implement this as a Canadian, Australian or Japanese investor, which I'm not. I will read that thing again more carefully.

One big problem is that I don't have any tax advantaged account, every transaction is a potential taxable event, but losses can be used to reduce gains, carried over indefinitely, and while the capital gains tax is about 25% (automatic withholding) my personal tax rate is most likely lower and I will eventually get some of the money back.

If anyone has any recommendations, links, ideas how to use this strategy in my situation that would be great.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Dual Momentum Investing
« Reply #720 on: August 28, 2015, 04:02:57 PM »

This is mostly right Sol.

One small point:  If today were your trading day, your lookback could be anything from 3 to 12 months and you would still move to short term treasuries.  This nicely illustrates why the particular lookback  period is not terribly important.  A significant downturn causes a surprisingly consistent signal regardless of the lookback you choose.
I disagree. I examined the performance of DM over the past 25 years, using a tool posted early in the thread, and your lookback period had a massive effect. The difference between a 3, 6 and 12 month lookback was considerable (with 3 month being the best performer, and 12 month the worst).

This difference, and the apparent arbitrary nature of it, is the reason that I initially chose not to employ this strategy.

Interesting claim. Please share your data.
I took the link to the resource you posted for me on page 5 and I ran a few tests. At first I was kinda impressed with the results, but after forummm pointed out lookback volatility I started screwing with lookback times one month at a time.

The short of it which I posted on page 9: Absolute momentum underperforms the S&P buy and hold back to 1983 if there is a 6 month loopback. However, it outperforms (CRAZY outperforms) with a 3 month loopback.

This led a discussion of "why" and the conclusion was "dunno".
« Last Edit: August 28, 2015, 04:05:40 PM by Chuck »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #721 on: August 28, 2015, 04:29:32 PM »
You must have made an error.

I just tested absolute momentum going back to 83 with 3 and 6 month lookbacks and ....

Both lookbacks beat the S&P. Both have smaller drawdowns.  Both have higher sharpe ratios.

Having difficulty sharing the images at work, sorry.

Again.  please share your data.
« Last Edit: August 28, 2015, 04:56:30 PM by milesdividendmd »

dungoofed

  • Pencil Stache
  • ****
  • Posts: 661
Re: Dual Momentum Investing
« Reply #722 on: August 28, 2015, 05:52:43 PM »
I know Mr. Antonacci has a post on his blog about how to implement this as a Canadian, Australian or Japanese investor, which I'm not. I will read that thing again more carefully.

Hi - any link for this blog post please?

hiddenace

  • 5 O'Clock Shadow
  • *
  • Posts: 10
Re: Dual Momentum Investing
« Reply #723 on: August 28, 2015, 07:41:54 PM »
I'm late to the party on this thread, but has anyone tried this theorymore actively on a much shorter time frame?

I mean momentum exists on an intra-day basis, so why not just look back at momentum on a 3-12 minute basis and apply this strategy to s&p futures?  Or a 3-12 second basis for that matter. Why are we arbitrarily choosing a long, multiple month time frame?

I guess what I'm asking is if this strategy is so good then why not apply it rapidly in day trading futures and truly maximize gains over a buy and hold?  Has anyone done it?   I may have to try it out on a paper account.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Dual Momentum Investing
« Reply #724 on: August 28, 2015, 07:56:02 PM »
I'm late to the party on this thread, but has anyone tried this theorymore actively on a much shorter time frame?

Yes, they're called high speed traders and they use supercomputers and billions of dollars to do it.  But good luck with your paper.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #725 on: August 28, 2015, 08:33:14 PM »

I'm late to the party on this thread, but has anyone tried this theorymore actively on a much shorter time frame?

I mean momentum exists on an intra-day basis, so why not just look back at momentum on a 3-12 minute basis and apply this strategy to s&p futures?  Or a 3-12 second basis for that matter. Why are we arbitrarily choosing a long, multiple month time frame?

I guess what I'm asking is if this strategy is so good then why not apply it rapidly in day trading futures and truly maximize gains over a buy and hold?  Has anyone done it?   I may have to try it out on a paper account.

There's not a lot of evidence that momentum  as it is defined here exists on a weekly or minute timeframe. In fact there's evidence that it doesn't.

There is plenty of evidence of mean reversion on short Time frames which is called "short term reversals."  This is the opposite of momentum. It is a bet that recent winners will start to lose as newinformation becomes digested by traders.

High speed traders are a different beast entirely. The premise (at least as described in flash boys) is built on front running publicly available information, and front running orders themselves. It's a form of temporal arbitrage. Not momentum trading.

Plus there is the problem of reliable price information. Haven't you noticed that different brokers will have different quotes for the same stock/etf at the same time?

Finally more trading means more trading costs and more taxes.

Not a winning formula in my book.

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #726 on: August 28, 2015, 11:24:39 PM »
I know Mr. Antonacci has a post on his blog about how to implement this as a Canadian, Australian or Japanese investor, which I'm not. I will read that thing again more carefully.

Hi - any link for this blog post please?

http://www.dualmomentum.net/2015/06/dual-momentum-for-non-us-investors.html

dungoofed

  • Pencil Stache
  • ****
  • Posts: 661
Re: Dual Momentum Investing
« Reply #727 on: August 28, 2015, 11:59:02 PM »
Thanks

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #728 on: August 29, 2015, 01:19:24 AM »

Hi, I'm a European investor, and I'm considering this strategy for part of my invested money.

Does anyone have recommendations for international investors? Is there anyone already using this strategy?
I'm specifically looking for info on which funds, ETFs or indices I should use, and which lookback periods seem appropriate. Any ideas? Since US and European markets are pretty correlated, does it make sense to e.g. choose S&P500 and EuroStoxx600 (plus some emerging markets for example)?
Due to currency valuations there was quite a different performance this year, so maybe the correlation isn't that big after all if one considers the currency exchange ratios.

I know Mr. Antonacci has a post on his blog about how to implement this as a Canadian, Australian or Japanese investor, which I'm not. I will read that thing again more carefully.

One big problem is that I don't have any tax advantaged account, every transaction is a potential taxable event, but losses can be used to reduce gains, carried over indefinitely, and while the capital gains tax is about 25% (automatic withholding) my personal tax rate is most likely lower and I will eventually get some of the money back.

If anyone has any recommendations, links, ideas how to use this strategy in my situation that would be great.

I know that Antonacci argues that dual momentum is extremely tax efficient since it tends to result in short-term capital losses and long-term capital gains.

Even in America, however, where I have a basic understanding of how the tax system works, I find it very difficult to model dual momentum after taxes. It is for this reason that I am strictly buy and hold in my taxable accounts,  and dual momentum in my retirement/tax-sheltered accounts.

As to the fund choices, for global equities momentum I would recommend either toggling between the S&P 500 and foreign developed markets, or US total market, and all world ex US. Any short-term high-quality treasury fund should work well for your cash position.

I hope that helps.


hiddenace

  • 5 O'Clock Shadow
  • *
  • Posts: 10
Re: Dual Momentum Investing
« Reply #729 on: August 29, 2015, 04:20:53 AM »
I'm late to the party on this thread, but has anyone tried this theorymore actively on a much shorter time frame?

Yes, they're called high speed traders and they use supercomputers and billions of dollars to do it.  But good luck with your paper.

I'll assume you weren't trying to sound snarky there, but I wasn't talking about HFT where companies use the fastest possible connection to take advantage of news events in fractions of a second.

I'm just saying if this method is using technical analysis to determine entry and exit points why not pick shorter time frames.

I didn't mention anything about competing with multi-billion dollar computers and fiber optics.

Can someone point me to the evidence that momentum doesn't exist intra-day? 

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #730 on: August 29, 2015, 11:39:38 AM »

I'm late to the party on this thread, but has anyone tried this theorymore actively on a much shorter time frame?

Yes, they're called high speed traders and they use supercomputers and billions of dollars to do it.  But good luck with your paper.

I'll assume you weren't trying to sound snarky there, but I wasn't talking about HFT where companies use the fastest possible connection to take advantage of news events in fractions of a second.

I'm just saying if this method is using technical analysis to determine entry and exit points why not pick shorter time frames.

I didn't mention anything about competing with multi-billion dollar computers and fiber optics.

Can someone point me to the evidence that momentum doesn't exist intra-day?

Proving the absence of something is always a tall order, and I would even go so far as to say that both momentum, and mean reversion probably do exist intraday.

But whether momentum trading exists as an actionable approach for the retail investor is another question entirely. After taxes, fees, friction, and accounting for imperfect price information I would guess not.

One thing is for certain.  momentum as originally defined by jagadheesh and titman and used as described in Dual Momentum, is a mid term effect existing for periods between 3 and 12 months.  In longer periods momentum is not predictive of future price movements and mean reversion (value investing) is.   

hiddenace

  • 5 O'Clock Shadow
  • *
  • Posts: 10
Re: Dual Momentum Investing
« Reply #731 on: August 29, 2015, 11:47:27 AM »
Ok fair enough, I suppose I should have asked for something highlighting the optimal length of time for implementing this strategy. In my naivete I assumed it could apply to shorter time frames like weeks, days, hours, or minutes instead of just months. I'm just trying to see why it only applies to months. It's a fascinating concept and I'm seeing if it can be applied in different scenarios.  But I guess it doesn't work that way...

dungoofed

  • Pencil Stache
  • ****
  • Posts: 661
Re: Dual Momentum Investing
« Reply #732 on: August 29, 2015, 09:09:27 PM »
I know Mr. Antonacci has a post on his blog about how to implement this as a Canadian, Australian or Japanese investor, which I'm not. I will read that thing again more carefully.

Hi - any link for this blog post please?

http://www.dualmomentum.net/2015/06/dual-momentum-for-non-us-investors.html

Actually I don't think that's much of an analysis at all as it misses out the most likely case ie investing in one's home economy, in one's home currency. I assume that is the portfolio you are trying to put together, right? ie
$HOMECOUNTRY Index Tracking ETF
ex-$HOMECOUNTRY Index Tracking ETF
$HOMECOUNTRY Treasury Bills ETF
$HOMECOUNTRY Bond ETF

Good luck with your back testing!

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #733 on: August 30, 2015, 12:48:52 PM »
I know that Antonacci argues that dual momentum is extremely tax efficient since it tends to result in short-term capital losses and long-term capital gains.

Unfortunately for me there is no difference in both short and long term gains, both are taxed the same. Used to be different, but not anymore.

Quote
As to the fund choices, for global equities momentum I would recommend either toggling between the S&P 500 and foreign developed markets, or US total market, and all world ex US.

Haven't found any ex-US funds yet, they seem to be pretty rare here. That's currently my biggest hurdle. I'd like to do this with locally available funds, and without having to trade abroad.

Quote
Any short-term high-quality treasury fund should work well for your cash position.

I think I'd go straight to cash. I get higher interest rates for cash than local short term treasuries.

Quote
I hope that helps.

Sure, thanks.

Actually I don't think that's much of an analysis at all as it misses out the most likely case ie investing in one's home economy, in one's home currency.

That is home bias. Very common on this forum with US investors overweighting US, but I don't really like home bias. Too risky long term (at least if you're a buy and hold investor).

Quote
I assume that is the portfolio you are trying to put together, right? ie
$HOMECOUNTRY Index Tracking ETF
ex-$HOMECOUNTRY Index Tracking ETF
$HOMECOUNTRY Treasury Bills ETF
$HOMECOUNTRY Bond ETF

No, If I wanted to do this it would be extremely easy. Ex-homecountry index ETF would simply be MSCI World since my home country is a very small part of the world economy. But I'd like to do Antonacci's original GEM.

I could modify GEM and use Eurostoxx600, and then  MSCI World ex EU. I can get these.

So if any other European investor has any ideas I'm all ears.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #734 on: August 30, 2015, 01:14:55 PM »
Can you list the funds available to you? Do you have  EAFE funds and S&P funds?

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #735 on: August 30, 2015, 03:30:49 PM »
Quote
Can you list the funds available to you?

Too many to list.
I trade on the local stock markets, www.justetf.com lists 1052 ETFs, among them 739 for equities and 242 for bonds, representing 668 different indices. I think I can trade most if not all of them. None from Vanguard btw.

Quote
Do you have  EAFE funds and S&P funds?

13 different ones for the pure S&P500, and another 13 for hedged, equal weight, volatility, max. dividend etc. variants.
No EAFE funds as far as I can see. I would have to use at least 3 or 4 different funds to build my own?


ChaseJuggler

  • Stubble
  • **
  • Posts: 123
Re: Dual Momentum Investing
« Reply #736 on: August 31, 2015, 09:55:32 AM »
Pulled the trigger! For the next 30 days, my portfolio is 100% bonds.


milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #737 on: August 31, 2015, 11:43:10 AM »
Me too.  trade goes through t close.

AZ

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Dual Momentum Investing
« Reply #738 on: August 31, 2015, 12:32:29 PM »
This is exciting. Real time proof of concept.

Was the method only tuned to satisfy past data, or is it a viable strategy that stands the test of a volatile market? LET'S FIND OUT

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #739 on: August 31, 2015, 12:37:59 PM »

This is exciting. Real time proof of concept.

Was the method only tuned to satisfy past data, or is it a viable strategy that stands the test of a volatile market? LET'S FIND OUT

In past data there were periods of underperformance lasting months or even years. So nothing can really be proved either way here other than the behavioral ease or difficulty of the approach.


Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Dual Momentum Investing
« Reply #740 on: August 31, 2015, 12:49:36 PM »

This is exciting. Real time proof of concept.

Was the method only tuned to satisfy past data, or is it a viable strategy that stands the test of a volatile market? LET'S FIND OUT

In past data there were periods of underperformance lasting months or even years. So nothing can really be proved either way here other than the behavioral ease or difficulty of the approach.
Well... That's not satisfying at all. Or conclusive.

I'm still gonna watch and pretend that it is.


milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #741 on: August 31, 2015, 01:49:15 PM »
Spend some time looking at this chart detailing month by month returns for GEM...  and yearly performance vs ACWI.


http://www.optimalmomentum.com/trackrecord3.html

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #742 on: August 31, 2015, 02:12:34 PM »
This is exciting. Real time proof of concept.

Was the method only tuned to satisfy past data, or is it a viable strategy that stands the test of a volatile market? LET'S FIND OUT

1 data point (1 month) won't tell you anything about a strategy. At all.

I too am in bonds in my retirement accounts, and mostly cash (SHY) in my trading account.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #743 on: September 01, 2015, 11:18:22 AM »
Its a weird feeling watching the market do poorly today, after having switched to short term treasuries yesterday.

It makes me feel smart, which is stupid.  I had no idea yesterday what the market would do today.  And I have no clue now what it will do tomorrow.  And of course the flip side of this feeling smart is that if the market had shot up today I would've felt stupid.

Logically I know that the DM approach is one that I am comfortable with, but I do not experience the market in real time with my deliberative "system 2" brain.  I experience it with my primordial "system 1" impulsive brain.

Which all helps to remind me that no matter what approach we choose, we are generally wiser to stick to our system through thick and thin...
« Last Edit: September 01, 2015, 11:45:36 AM by milesdividendmd »

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1463
  • Age: 44
  • Location: Texas
Re: Dual Momentum Investing
« Reply #744 on: September 01, 2015, 11:40:39 AM »
A little off topic but I think you mean System 1 is your impulsive brain and System 2 is your logical brain. At least that's what I read in Thinking Fast and Slow. :)

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #745 on: September 01, 2015, 11:45:56 AM »
A little off topic but I think you mean System 1 is your impulsive brain and System 2 is your logical brain. At least that's what I read in Thinking Fast and Slow. :)

Thanks!  and amended.

starguru

  • Pencil Stache
  • ****
  • Posts: 752
Re: Dual Momentum Investing
« Reply #746 on: September 01, 2015, 11:59:17 AM »
Its a weird feeling watching the market do poorly today, after having switched to short term treasuries yesterday.

It makes me feel smart, which is stupid.  I had no idea yesterday what the market would do today.  And I have no clue now what it will do tomorrow.  And of course the flip side of this feeling smart is that if the market had shot up today I would've felt stupid.

Logically I know that the DM approach is one that I am comfortable with, but I do not experience the market in real time with my deliberative "system 2" brain.  I experience it with my primordial "system 1" impulsive brain.

Which all helps to remind me that no matter what approach we choose, we are generally wiser to stick to our system through thick and thin...

I thought the procedure was to look AFTER the last trading of the month, and then the next trading day to move to the indicated asset -- so you should be selling today, not yesterday.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #747 on: September 01, 2015, 12:12:47 PM »
I trade on the turn of the month, so my trade went through after close yesterday.  As you can see from this thread, different people have different trigger dates.  Trading a week early, like others did would have been even "smarter," in this case (thus far).  But long term, this random variability shouldn't make a great deal of difference.

yoda34

  • 5 O'Clock Shadow
  • *
  • Posts: 55
Re: Dual Momentum Investing
« Reply #748 on: September 04, 2015, 10:42:34 AM »
Hi all - long time lurker, not a prolific poster. I've followed this thread with interest given that I have some experience in this field. I have a few thoughts about active investing in general (not just dual momentum) and this seemed to be the current thread to have that discussion. My apologies in advance for the dense post - had a lot of thoughts over the past few weeks to get out.

First there are several identified market anomalies that do return more than a standard market index (S&P500, Total Market, what ever). These have been identified, studied and acknowledged through mountains of academic research. Two of those anomalies are value and momentum. These were first widely identified by fama and french (ironically enough the fathers of efficient market theory). Of course, the outstanding question is do these anomalies really return more on a risk adjusted basis (i.e. do you get more overall gain due to risk) but it depends on how you measure risk. Modern portfolio theory will measure volatility and standard deviation and declare that value and momentum do not beat the market on a risk-adjusted basis. Value investors will say that's a crazy way to measure risk (risk being more identified with "overpaying" for a security and a "margin of safety" from Ben Graham). Regardless it is not difficult at all to construct portfolios that exploit value or momentum (as dual momentum does) to reliably return absolute gains greater than any market index over long periods of time. I can give you three value methods off the top of my head that do so.

The question is then "If it is easy and provable to build a system based on value and momentum that returns absolute gains greater than the market, why doesn't everyone beat the market?"

First let's discuss index buy and hold approaches. First, if you believe in a strong efficient market theory (i.e. the market price is never wrong) then buy and hold is not only smart, it's literally the only approach that makes sense. You receive the average market return and assume the average market risk which is the best you can ever do. Even if you don't believe in totally strong efficient market, buy and hold could still make sense as you may be happy with receiving the mean return at the mean risk with very little work or stress.

But if you believe, as I do and as I believe research has shown, that there are at least two anomalies that return more than the market in value and momentum why is it so hard to take advantage of?

1. Behavioral bias errors
2. Frictional costs (especially in taxable accounts)

Behavioral bias errors refer to the fact that, as humans, we are incredibly poor at making informed decisions. Our decision making process is heuristic and pattern based and is incredibly fast. It had to be in order to survive through millions of years of evolution. Unfortunately that process also makes us extremely bad at making decisions that require non emotional thinking and we often fool ourselves by seeing patterns where none exist (thanks brain!). Joel Greenblatt (of MFI fame) did a 10 year study where he found that investors systematically avoid stocks with large returns and panic and sell during down turns at exactly the wrong times - repeatedly. Unless you can be extremely disciplined you can not make active investing work. Even knowing that Value approaches beat the index - can you ride a 50% drawdown or 30% standard deviation for 15 years? Because that's exactly what it takes. Most retail investors can't. Professionals have a short time bias problem due to having to keep their jobs and so they can't. In general people actively destroy any excess returns (and them some) through these errors.

Frictional costs are also a HUGE problem. Most of the value or momentum strategies that generate the large excess returns over the market (think 20% per anum) require trading at intervals of less than 1 year. Even putting aside the transaction costs for each trade the difference between the long term cap gains and the short term cap gains is enormous. If you assume a long term cap gain tax of 23.8% and a short term of 36.8 - 42.8% (the top two highest tax brackets) then the following is true: Assuming a market return of 8% per year you would have to earn 14% just to break even on the taxes (assuming that you continued to hold the index and didn't willy nillly sell). So you not only have to beat the market, you have to beat the market by 75% to just break even if your strategy causes you to incur short term capital gains!!! Also, the short term capital gains hits are compounded right along with your gains meaning over a 15 year period the difference in long term taxes and short term taxes can literally be 1000s of percents if you don't make at least that 75% premium to break even. This is extremely hard.

My personal opinion is this is why Warren Buffet is so successful. He buys using a value strategy (known anomaly to market returns) but then NEVER sells reducing not only his taxes but completely eliminating all behavioral bias errors. He doesn't mess up because he refuses to play the game.

Ok - so enough is enough. What does this all mean. Can dual momentum beat the a buy and hold index. Yes it certainly can. IF you (a) are a super iron man on discipline and never make behavioral mistakes and (b) find a momentum system that either only makes long term capital trades or beats the market by WIDE margin to make up the difference.

I think you'll find the reality of actually executing A and B very hard in real life, which is why for almost everyone a buy and hold index strategy will far and away be the superior choice. I would say just my 2 cents, but this post has to be way more than that. Sorry for the rambling and thanks for reading.
« Last Edit: September 04, 2015, 11:23:12 AM by yoda34 »

kvaruni

  • 5 O'Clock Shadow
  • *
  • Posts: 92
  • Location: Northern Ireland, UK
Re: Dual Momentum Investing
« Reply #749 on: September 04, 2015, 11:31:03 AM »
Wow, long live the long time lurkers. This is a very clear post, and it probably made a lot more clear to me about dual momentum investing than I ever understood from just gleaning over all other posts on this topic. Kudos to you yoda34!

 

Wow, a phone plan for fifteen bucks!