Author Topic: Dual Momentum Investing  (Read 1943045 times)

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #900 on: October 25, 2015, 04:48:41 AM »

So we'll just tragedy of the commons away the problem?

No idea what you are trying to say here.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #901 on: October 25, 2015, 04:50:29 AM »

Aren't there much bigger problems in the markets that we let happen?

Please milesmd, admit defeat so we can  discuss more important stuff.

Why on earth would I admit defeat when the other side of the argument is clearly failing to make a cogent point?

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #902 on: October 25, 2015, 05:17:40 AM »
> Why on earth would I admit defeat

So we can move on.

> when the other side of the argument is clearly failing to make a cogent point?

Because they will hopefully stop. Otherwise this fruitless discussion will go on for another 19 pages. Waste of time and energy.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #903 on: October 25, 2015, 05:46:26 AM »
I tend to agree generally with your definition that "volatility is just short term momentum", but I would be more specific.

Volatility is instantaneous momentum.

Why?  This definition would make sense only if you are measuring volatility over correspondingly short time spans.  If you zoom out in your perspective, the "short term" momentum that matters gets correspondingly longer.

If you chart continuous price movements, intraday prices could gyrate wildly even if the closing price ends up exactly where it started.  On a chart plotting daily prices at market close, that intraday volatility disappears--the price line from one day to the next becomes a straight line rather than an oscillating zig-zag.  And the same is true with respect to intraweek volatility on a chart plotting only week-end prices, and so on.

Quote
What has not been successfully argued yet by you or anyone, is that there is any relationship between the price momentum of a specific look back period between 3 and 12 months and the price movement at the moment that a DM trader places his trade.

This the fundamental disconnect we've been having.  If you're talking about volatility over periods of time equaling the momentum trader's lookback period, then we've already established that we all agree that momentum trading increases volatility (since momentum trading axiomatically amplifies any price swings occurring over those periods).  We don't need to demonstrate any relationship between the price trend over the lookback period and the price trend over the moments before the trade is executed, because we haven't been arguing that momentum trading increases second-by-second volatility (though, in a microcosm of the phenomenon that nicely illustrates our point, a momentum trader using a one-second lookback period would do exactly that (and in that scenario, a micro version of milesdividendmd would argue that no, this micro-momentum trader is not amplifying volatility, because the momentum trade does not necessarily follow the trend in effect in the nanoseconds leading up to its execution)).

so we can discuss more important stuff.

Who's stopping you?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #904 on: October 25, 2015, 09:09:20 AM »

I tend to agree generally with your definition that "volatility is just short term momentum", but I would be more specific.

Volatility is instantaneous momentum.

Why?  This definition would make sense only if you are measuring volatility over correspondingly short time spans.  If you zoom out in your perspective, the "short term" momentum that matters gets correspondingly longer.

If you chart continuous price movements, intraday prices could gyrate wildly even if the closing price ends up exactly where it started.  On a chart plotting daily prices at market close, that intraday volatility disappears--the price line from one day to the next becomes a straight line rather than an oscillating zig-zag.  And the same is true with respect to intraweek volatility on a chart plotting only week-end prices, and so on.

Quote
What has not been successfully argued yet by you or anyone, is that there is any relationship between the price momentum of a specific look back period between 3 and 12 months and the price movement at the moment that a DM trader places his trade.

This the fundamental disconnect we've been having.  If you're talking about volatility over periods of time equaling the momentum trader's lookback period, then we've already established that we all agree that momentum trading increases volatility (since momentum trading axiomatically amplifies any price swings occurring over those periods).  We don't need to demonstrate any relationship between the price trend over the lookback period and the price trend over the moments before the trade is executed, because we haven't been arguing that momentum trading increases second-by-second volatility (though, in a microcosm of the phenomenon that nicely illustrates our point, a momentum trader using a one-second lookback period would do exactly that (and in that scenario, a micro version of milesdividendmd would argue that no, this micro-momentum trader is not amplifying volatility, because the momentum trade does not necessarily follow the trend in effect in the nanoseconds leading up to its execution)).

so we can discuss more important stuff.

Who's stopping you?

Now we're getting somewhere Brooklyn. Well argued.

The reason I (and Sol) argue that volatility is (very) short term momentum is simply because I (we?) believe this definition to be true.

When people talk about the market being "volatile" or "unstable" they are almost universally talking about large and fast price swings that are readily apparent to market participants in real time.

Do you deny that...

A. This is the "volatility" which Sol was describing in his original critique?

And

B. This is the dominant usage of the phrase "market volatility?"

When the market drops 2 percent and then reverses course and rises 4 percent, we would all describe that as "volatile."

When the market rises by a compounding 7% over the course of decades, that slow course is really never described as "volatile" though the intraday fluctuations invisible on the long term chart and underlying that long term trend may well be.

More to the point, if you honestly believe that long term price movements are "volatility," then wouldn't you concede that buy and holders, whose only possible trades in the accumulation phase are to buy more securities contribute more than DM traders (who after all occasionally exit their positions) to "volatility?"

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #905 on: October 25, 2015, 02:10:47 PM »
A. This is the "volatility" which Sol was describing in his original critique?

Yes, I think the "volatility" that Sol referred to in his original critique of momentum trading (and the "volatility" the rest of us have been referring to in our subsequent posts that built upon that original critique) is the same "volatility" that necessarily gets amplified by momentum trading, as I think the content of his post made clear (and which is why some of us were having such difficulty understanding your resistance to that critique).

Quote
B. This is the dominant usage of the phrase "market volatility?"

I think the measurement period people have in mind when they generally speak of the "volatility" of the stock market is not any precise time span in particular but just a vague, loosely-defined notion of "the short term," which has no bright-line cut-off but which generally includes multi-month and multi-year periods (long enough to contain the market swings that get amplified by momentum trading strategies that use lookback periods in the 3-12 month range) but would not include, say, multi-century periods.  For example, owing to the stock market's "volatility," people often caution against investing in it if your time horizon is less than "five years" or "several years" as a general rule of thumb.

Quote
More to the point, if you honestly believe that long term price movements are "volatility," then wouldn't you concede that buy and holders, whose only possible trades in the accumulation phase are to buy more securities contribute more than DM traders (who after all occasionally exit their positions) to "volatility?"

As I've said, buy and hold investing puts upward pressure on prices and therefore does contribute to the long-term upward trend of the market.  Any investing strategy that involves the purchase or sale of investments (which is to say, every investing strategy) puts either upward or downward pressure on the price of those investments at the time of the purchase or sale and, in that sense, "contributes to volatility" in one way or another.  What is distinct about momentum trading is that price itself serves as the trading signal.  B&H investing pushes the market up, while momentum trading pushes the market in whatever direction it happens to have moved over the applicable lookback period.  And the shorter the lookback period is, the "shorter term" the volatility-amplification is.  Versions of momentum trading that use a one-second, one-month or one-year lookback period can all be said to amplify "volatility" consistent with general usage of the term (that is, volatility over time horizons that generally matter to investors in the stock market).  A version of momentum trading that uses a one-century lookback period, on the other hand, can only be said to amplify volatility over centuries-long time horizons, which is not consistent with general usage of the term because it is (at least at the present time) beyond the concern of virtually all investors actually participating in the market.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #906 on: October 25, 2015, 05:38:50 PM »
Brooklyn,

I have literally never heard anyone use volatility as you have defined it, unless they were incorrectly conflating it with another word.

For instance the VIX volatility index does not seek to measure the price movement of the stock market in the timeframe of years or months. It seeks to measure rapid intraday price swings, reflective of investor fear.

Another example to illustrate how your definition of "volatility" is in no way credible.

I have literally never heard someone say anything along the lines of , "the S&P has risen 12% points in the last six month, whereas usually it only rises 3.5%. The stock market sure is volatile."

Since Sol himself defined volatility as "short term momentum" I find it quite apparent that when he describes "volatility" and "unstable markets" he is not talking about price movements of months or years.

Furthermore look at how Sol makes his recent argument here....

"Volatility is just short term momentum.  Lots of momentum traders, trading on different dates in the same direction that the market is moving on that day, will increase short term momentum, which is volatility.  The next day if the market direction reverses for some reason unrelated to momentum traders, all of the momentum traders who trade that day will trade again in the new direction, increasing short term momentum again.  Two days, two different groups of momentum traders with different trade dates, amplifying the market gyrations.  Creating volatility."

Note how he specifies, that the momentum traders are making trades in the same direction of market price movement "ON THAT DAY."  Notice how focused he is on readily apparent "market gyrations."

He is and was always talking about fast changes in price (as in flash crashes.) This likely is why he originally claimed that DM led to market "instability."

So you can claim that you all simultaneously were using "volatility" in a manner in which literally no one else uses it including the person who brought it up here originally, but Occam's razor tells me that the more likely explanation that you were all conflating "volatility" with "momentum."

An easy mistake to make, I've improperly used words before and admitted it to you on this very forum, and probably this very thread.

We all use language imprecisely at times so instead of doubling down with a novel and absurd definition of "volatility",  the wiser course is probably to just admit you are human, subject to errors of language, and move on.
« Last Edit: October 25, 2015, 05:41:11 PM by milesdividendmd »

MDM

  • Senior Mustachian
  • ********
  • Posts: 11473
Re: Dual Momentum Investing
« Reply #907 on: October 25, 2015, 06:52:26 PM »
I have literally never heard anyone use volatility as you have defined it, unless they were incorrectly conflating it with another word.
Reading only this much, I went searching for how VIX is calculated.

Then came back here, only to see in the very next line:
Quote
For instance the VIX volatility index does not seek to measure the price movement of the stock market in the timeframe of years or months. It seeks to measure rapid intraday price swings, reflective of investor fear.
As noted above, my "in depth" VIX knowledge is limited to a quick internet search, and we all know how (un)reliable Wikipedia can be, etc.  But in https://en.wikipedia.org/wiki/VIX it does say
Quote
The VIX is quoted in percentage points and translates, roughly, to the expected movement in the S&P 500 index over the upcoming 30-day period, which is then annualized. "VIX" is a registered trademark of the CBOE.
Is it then correct to say that the VIX measures monthly volatility?

And there is also https://en.wikipedia.org/wiki/Volatility_%28finance%29 - haven't read this thread in enough detail to know "whose side" that supports....

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #908 on: October 25, 2015, 07:07:33 PM »

I have literally never heard anyone use volatility as you have defined it, unless they were incorrectly conflating it with another word.
Reading only this much, I went searching for how VIX is calculated.

Then came back here, only to see in the very next line:
Quote
For instance the VIX volatility index does not seek to measure the price movement of the stock market in the timeframe of years or months. It seeks to measure rapid intraday price swings, reflective of investor fear.
As noted above, my "in depth" VIX knowledge is limited to a quick internet search, and we all know how (un)reliable Wikipedia can be, etc.  But in https://en.wikipedia.org/wiki/VIX it does say
Quote
The VIX is quoted in percentage points and translates, roughly, to the expected movement in the S&P 500 index over the upcoming 30-day period, which is then annualized. "VIX" is a registered trademark of the CBOE.
Is it then correct to say that the VIX measures monthly volatility?

And there is also https://en.wikipedia.org/wiki/Volatility_%28finance%29 - haven't read this thread in enough detail to know "whose side" that supports....


Uh oh, another semantic sidetrack, and I introduced it this time...

In all honestly I  believe it's a touch more meta than your definition.

My understanding is that it measures investors current expectations for future volatility in the coming 30 day period.

It is of course commonly referred to as "the fear index."

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #909 on: October 25, 2015, 10:13:35 PM »
So you can claim that you all simultaneously were using "volatility" in a manner in which literally no one else uses it including the person who brought it up here originally, but Occam's razor tells me that the more likely explanation that you were all conflating "volatility" with "momentum."

Unlike you, I don't claim to be able to see inside sol's mind.  I can only tell you how I interpreted his posts based on the words they contained.  And I can tell you with certainty what I intended with my own posts.  In both cases, it was that momentum trading strategies amplify the very market movements on which they trade.  Given that (i) you alone have been resisting this argument as expressed by him and me, (ii) other posters have chimed in to express confusion about your resistance, and (iii) according to the dictionary, there is nothing incorrect (let alone absurd) about my definition of "volatility," Occam's razor tells me that you've got things exactly backwards.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #910 on: October 25, 2015, 11:06:33 PM »
I never claimed to be able to see into Sol's mind. But I can certainly read his words.

Since he defines "volatility" in writing in his own words as "short term momentum," and he takes pains to point out how a DM trader could make a trade in the same direction of price movement on the day he trades thereby increasing "volatility," how can you possibly imagine that he is using "volatility" to mean anything other than short term momentum?

That's a question I have now asked you many times in many different forms. Why not answer it?
« Last Edit: October 25, 2015, 11:53:30 PM by milesdividendmd »

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Dual Momentum Investing
« Reply #911 on: October 26, 2015, 12:28:02 AM »

Brooklyn,

I have literally never heard anyone use volatility as you have defined it, unless they were incorrectly conflating it with another word.

For instance the VIX volatility index does not seek to measure the price movement of the stock market in the timeframe of years or months. It seeks to measure rapid intraday price swings, reflective of investor fear.

Another example to illustrate how your definition of "volatility" is in no way credible.

I have literally never heard someone say anything along the lines of , "the S&P has risen 12% points in the last six month, whereas usually it only rises 3.5%. The stock market sure is volatile."


Wait, wait.. You're saying volatility is only daily? 

So if the market is up 14% over a two week period (1% per day), and down 14% the next two weeks, then back up, etc. you'd say that it wasn't volatile, because it was only 1% per day?

I'd sure as heck call that volatile.  My definition would be much closer to BG's than the super narrow "daily"--yes, I think we could have a volatile market over months.

You may not say it's volatile if it's only gone up over the last six months, but if it's spiked up, and down, and up, peak to trough several times, yes, absolutely I'd say "the market's been volatile over the last six months" even if there were 0 daily huge swings, or flash crashes.

To further prove the point: have you ever heard someone say "short term volatility"?  If so, and it made sense, clearly there's volatility over a longer term. Otherwise that phrase is redundant nonsense.

Short term volatility is the daily, like flash crashes. We are talking about over weeks, or months. Or, as you call it, momentum.

Quote
So you can claim that you all simultaneously were using "volatility" in a manner in which literally no one else uses it including the person who brought it up here originally, but Occam's razor tells me that the more likely explanation that you were all conflating "volatility" with "momentum."

Wouldn't Occam's razor not be that we were all simultaneously using it incorrectly, but rather than your definition was incorrect and we were all using it just fine?

In any case, it appears now we have the heart of the disagreement.  We're saying momentum trading leads to volatility, rather than stability, in the markets over various time periods. Not necessarily on a single day, leading to a flash crash (though it might do that too), but in general in the markets.
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.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Dual Momentum Investing
« Reply #912 on: October 26, 2015, 12:29:45 AM »

I never claimed to be able to see into Sol's mind. But I can certainly read his words.

Since he defines "volatility" in writing in his own words as "short term momentum," and he takes pains to point out how a DM trader could make a trade in the same direction of price movement on the day he trades thereby increasing "volatility," how can you possibly imagine that he is using "volatility" to mean anything other than short term momentum?

That's a question I have now asked you many times in many different forms. Why not answer it?

Probably because your "short term" is different than ours. I don't know what sol's is, but when I read his quote, I'm thinking days, weeks, or even months. You're apparently thinking hours, or maybe a single day.

We're reading the same quote, but with different definitions of short term.
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
Re: Dual Momentum Investing
« Reply #913 on: October 26, 2015, 12:56:53 AM »


I never claimed to be able to see into Sol's mind. But I can certainly read his words.

Since he defines "volatility" in writing in his own words as "short term momentum," and he takes pains to point out how a DM trader could make a trade in the same direction of price movement on the day he trades thereby increasing "volatility," how can you possibly imagine that he is using "volatility" to mean anything other than short term momentum?

That's a question I have now asked you many times in many different forms. Why not answer it?

Probably because your "short term" is different than ours. I don't know what sol's is, but when I read his quote, I'm thinking days, weeks, or even months. You're apparently thinking hours, or maybe a single day.

We're reading the same quote, but with different definitions of short term.

Then why does Sol bother to specify that a DM trader could increase volatility by making a trade in the same direction as the price movement of the day that he trades?  If Sol is thinking "weeks or even months" then what is the possible significance of the price direction on the day the trader trades?

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #914 on: October 26, 2015, 01:20:53 AM »
And the problem with your's and Brooklyn's definition of market volatility, as opposed to Sol's, by the way, is that it is so non specific as to be rendered meaningless.

If a trader buys an S&P index fund in a moment when it's price is going up on a day when it's price is going down, during a week when it's price is going up, during a month when it's price is going down, during a six month period when it's price is going up, then is he increasing or decreasing volatility?

Do you see the problem there? How can you argue DM's effect either way if you can't even define whether a simple trade increases or decreases volatility?
« Last Edit: October 26, 2015, 01:22:35 AM by milesdividendmd »

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #915 on: October 26, 2015, 10:25:42 AM »
I never claimed to be able to see into Sol's mind. But I can certainly read his words.

Since he defines "volatility" in writing in his own words as "short term momentum," and he takes pains to point out how a DM trader could make a trade in the same direction of price movement on the day he trades thereby increasing "volatility," how can you possibly imagine that he is using "volatility" to mean anything other than short term momentum?

In that particular post (post # 896), I thought sol was saying that a momentum trader who trades on a single-day price swing (i.e., using a one-day lookback period) would amplify day-by-day volatility.  Why else would there be traders necessarily trading with the new direction of the market when it shifted directions on day two?

Having now reread that post, I see that he was not talking about momentum traders re-entering the market because of the single-day market movement, but a different group of momentum traders executing their trades on day two.  So, like you, I don't understand what sol's point was in that post.  If the day-two traders are not using a single-day lookback period, their trades would not necessarily be in the same direction that the market moved over that single day.

I'd be interested in hearing sol's explanation of what point he was trying to make in that post, because I had been interpreting all of his previous posts on this subject as arguing that momentum trading is self-amplifying in that the piling on to market gyrations will amplify those gyrations (which is what his posts had been saying).  Given that sol seems to have less appetite for continuing to engage in this debate than you and I do, we may not get the benefit of hearing sol explain his own posts.  In any event, I would suggest that we stop engaging in the Talmudic construction of the Words of Sol and focus instead on the argument that I myself am actually making (and have been making), whether or not it is identical to the argument sol has been making.

And the problem with...Brooklyn's definition of market volatility...is that it is so non specific as to be rendered meaningless.

My definition of volatility is no different than your definition, except that I'm saying it can occur over any given time period rather than only extremely short-term time periods.  In the following chart, I would say MilesDividendMD, Inc. has higher volatility than BrooklynGuy Corp.:



And that is true without knowing the specific time intervals for the y-axis.  Whether the chart is displaying prices over a one-minute time span or a one-decade time span, one asset has high volatility in relation to the other for the period in question.

You say that you have literally never heard anyone speak of the stock market's volatility over a measurement period extending over months or years and find it absurd that I'm using the word "volatility" to describe market gyrations over month- or year-long periods.  I strongly disagree that my usage of the word is any way incorrect (let alone absurd) or inconsistent with general usage of the word as it pertains to the stock market (again, consider my example of widespread disclaimers against using the stock market as a wealth-building tool for time horizons less than five years or so due to the stock market's volatility, not to mention the use of the term in sources like those MDM cited above), but let's just put aside this purely semantic argument because it is a distraction from the central debate.

Quote
If a trader buys an S&P index fund in a moment when it's price is going up on a day when it's price is going down, during a week when it's price is going up, during a month when it's price is going down, during a six month period when it's price is going up, then is he increasing or decreasing volatility?

Do you see the problem there? How can you argue DM's effect either way if you can't even define whether a simple trade increases or decreases volatility?

The only "problem there" is that in your question you are using "volatility" as an absolute term rather than a relative term--the word has no meaning except in reference to a given measurement period.  It's no different than asking whether your hypothetical trader is following a trend or diverging from a trend (or adding to momentum or detracting from momentum).  Depending on the period in question, the answer differs.  You are laser-focused on what you have called "instantaneous momentum"--the volatility occurring in the instant leading up to the trade--which you seem to believe is the only measurement period for volatility that can legitimately be called "short term" and can therefore legitimately be called "volatility" at all.  Again, a micro-version of the argument you are making would dispute that even second-by-second market gyrations qualify as "volatility," because the trends occurring over periods as long as a full second are way too "long-term" to matter to an investor who is only focused on trends occurring over billionths of a second.

Here, once again, is my argument in a nutshell (which thankfully did not use the word "volatility"):

Momentum trading involves identifying a pricing trend, over some period, that is already underway, and then following that trend.  The momentum trader executes a trade in the same direction as the trend.  That trade has the effect of amplifying the trend.  The amplification effect may be minimal (and, in the case of a typical trade executed by individual trader like yourself, infinitesimal), and the activity of other traders may counteract the amplification effect, but that is the effect.  It has to be, as a matter of logic.  And the bigger the share of the market momentum traders make up, the stronger the overall effect will be.

You agree with this argument because, as you have said, it is axiomatically correct.  Momentum trading contributes to the amplification of the very market gyrations that lead the momentum trader to make his trades.  If the lookback period is three months, market gyrations occurring over three month periods will be amplified.  If the lookback period is twelve months, market gyrations occurring over twelve month periods will be amplified.  If you don't want to call market gyrations over these time periods "short term volatility," so be it.  Let's call them "medium term volatility" or "long term volatility."  The label does not matter.  But the substance of the argument is that momentum trading using a 3-12 month lookback period amplifies market gyrations over corresponding periods, in a manner totally unrelated to market fundamentals.  Whether or not you want to call this effect on 3-12 month market gyrations an effect on "volatility," I find it to be disruptive to the market in a way that, say, amplification of market gyrations over 100-year time spans would not be.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #916 on: October 26, 2015, 10:59:27 AM »
Every trade including buy and hold amplifies some gyrations and diminishes others depending on the time period that you look at. In this sense every trade both increases and decreases volatility. To make a claim that one strategy increases volatility over another you really have to specify a time frame.

I will agree that in your graph md2 stock is more volatile than Brooklyn stock. But I would simply point out that this use of "volatile" is a very different usage of the word "volatile" from when someone says "of late there has been increased market volatility."  Volatility used in this manner is used to denote unusually rapid high amplitude price movements.

Importantly, in Sol's claim that DM increased volatility and market instability he was almost assuredly using "volatility" in this second meaning of the word, based on his own words and context.

Do a google search for "increased market volatility" and you will find that the articles that come are about rapid high amplitude price movements that are readily apparent to market participants in real time. You will also find that these articles pop up at times of market tumult as is September 2015.

Importantly what you have not demonstrated, is that the DM strategy increases the likelihood that the stock market will behave more like my stock and less like yours.

That's a different argument, but one that I am also happy to engage in.

If you like, please make the argument for why you feel that DM is more likely to lead to "volatility" in this second definition of the word.

I am dubious.

And what's wrong with getting "Talmudic", by the way?

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #917 on: October 26, 2015, 11:10:37 AM »
I'm afraid I've neglected my day job in favor of participating in the forum enough for one day so I'm going to have to wait until I have more time to give a full response, but I want to respond to this bit now just in case you thought I might have been implying something nefarious:

And what's wrong with getting "Talmudic", by the way?

All I meant is that we were starting to dissect sol's posts to discern their meaning (the way a Talmudic scholar would the Talmud) instead of debating the merits of the argument I am actually making (regardless of whether or not that argument is consistent with the argument sol was actually making).

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #918 on: October 26, 2015, 11:33:31 AM »
I don't believe 2 sides can actually debate the merits of an argument unless they can agree on what they are arguing about. That's the genius of Talmudic thinking. (I imagine you engage in this sort of thinking  every day in your day job.)

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #919 on: October 26, 2015, 11:36:41 AM »
And the importance of Sol's original argument is that he is the one who made the original claim. So that is what we should have been arguing about, until we chose to move on to a new point of disagreement.
« Last Edit: October 26, 2015, 11:44:04 AM by milesdividendmd »

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #920 on: October 26, 2015, 02:06:50 PM »
Ok, I couldn't help myself -- I'm back.  My name is brooklynguy and I am a forum addict...

I don't believe 2 sides can actually debate the merits of an argument unless they can agree on what they are arguing about. That's the genius of Talmudic thinking. (I imagine you engage in this sort of thinking  every day in your day job.)

And the importance of Sol's original argument is that he is the one who made the original claim. So that is what we should have been arguing about, until we chose to move on to a new point of disagreement.

Fair enough.  I thought we had already established that you and I, at least, were arguing about different things, because we interpreted sol's original claim differently.  Rebs and frugalnacho seem to have understood "our side" of the debate in the same way I did.  If sol decides to reengage and shed light on his understanding of his own argument, then we'll get more insight into whether or not he did too.  But now that we recognize that we in fact did not agree on what we were arguing about, shouldn't we move on to argue over the point of disagreement about which we (hopefully) do agree that we indeed disagree?

Every trade including buy and hold amplifies some gyrations and diminishes others depending on the time period that you look at. In this sense every trade both increases and decreases volatility. To make a claim that one strategy increases volatility over another you really have to specify a time frame.

Yes, I agree -- this was exactly my point.

Quote
I will agree that in your graph md2 stock is more volatile than Brooklyn stock. But I would simply point out that this use of "volatile" is a very different usage of the word "volatile" from when someone says "of late there has been increased market volatility."  Volatility used in this manner is used to denote unusually rapid high amplitude price movements.

Quote
Do a google search for "increased market volatility" and you will find that the articles that come are about rapid high amplitude price movements that are readily apparent to market participants in real time. You will also find that these articles pop up at times of market tumult as is September 2015.

It's a different usage only in the time period in question, right?  If "of late" means "in the last few days," then it's referring to market gyrations over the space of the last few days.  If "of late" means "the last few months," then it's referring to market gyrations over the space of the last few months.

Again, it's no different than usage of the phrase "of late the market has trended up."  If "of late" means the last few days, it's referring to an upward trend over the last few days.  If "of late" means the last few months, it's referring to an upward trend over the last few months.

I definitely agree with you that when people generically refer to market volatility (without specifying the precise time period to which they are referring), they are more likely to be referring to time frames on the shorter end of the spectrum.  Price swings over the space of a day or a week or a month get much more attention in the financial press and at cocktail parties than price swings occurring over longer periods .  But I do not agree that price swings occurring over the course of a year or even several years cannot be, and never are, described as market volatility.

Quote
Importantly, in Sol's claim that DM increased volatility and market instability he was almost assuredly using "volatility" in this second meaning of the word, based on his own words and context.

Is the "second meaning" what I just described above?  The same as the first meaning, except as to the relevant time span?  Market swings occurring over periods short enough to qualify for some unspecified cutoff, but not market swings occurring over longer periods?

My reading of sol's original claim, and all of his subsequent posts except for the one you cited above (post # 896, which I am confused by), is that the "volatility" he was referring to is the same "volatility" that actually gets amplified by momentum trading--namely, the market gyrations that occur over whatever lookback period the momentum traders are using.  That is was he repeatedly claimed, isn't it?  That piling on to every price swing magnifies the amplitude of those price swings?

If that is not what sol intended to claim, so be it, but that is what I am claiming and have been claiming.

Quote
If you like, please make the argument for why you feel that DM is more likely to lead to "volatility" in this second definition of the word.

If I'm understanding you correctly, here you mean the argument that momentum trading using a lookback period in the 3-12 month range contributes to volatility over shorter periods (days? weeks?) (that is, the argument I understand that you think/thought sol was making, and that you previously thought I was making).  I would not make that argument, because I don't think it is true.  I think momentum trading contributes to volatility over periods containing subperiods equal to the relevant lookback period, not shorter periods.

Quote
Importantly what you have not demonstrated, is that the DM strategy increases the likelihood that the stock market will behave more like my stock and less like yours.

That's a different argument, but one that I am also happy to engage in.

If the time period covered by the chart above is long enough to include multiple lookback periods, then momentum trading necessarily causes the market to behave more like your stock than my stock because (as I think we all agree) momentum trading has the self-reinforcing effect of amplifying the price trends it trades on.  Momentum traders will pile on to every price swing occurring over the lookback periods contained within the overall time period covered by the chart and thereby amplify those price swings.  Let's suppose the chart covers a two year period, and we're talking about momentum traders who use a three-month lookback.  At the end of three months, if the market has trended up, those traders will pile on (by purchasing shares) and thereby push the market further up.  Over the next three months, if the market has trended down, they will pile on (by selling shares) and thereby push the market further down.

Now, as I believe everyone on all sides of this debate has explicitly recognized, this effect requires sufficient presence of momentum trading in the market in order to be meaningful and it can be counteracted by other forces in the market.  But the greater the share of the market that momentum traders make up, the stronger the gyration-amplification effect will be, and the less susceptible it will be to counteraction by other market participants.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #921 on: October 26, 2015, 03:51:54 PM »
Ok, I couldn't help myself -- I'm back.  My name is brooklynguy and I am a forum addict...

It's an epidemic, and I've got it bad.

Quote
If the time period covered by the chart above is long enough to include multiple lookback periods, then momentum trading necessarily causes the market to behave more like your stock than my stock because (as I think we all agree) momentum trading has the self-reinforcing effect of amplifying the price trends it trades on.  Momentum traders will pile on to every price swing occurring over the lookback periods contained within the overall time period covered by the chart and thereby amplify those price swings.  Let's suppose the chart covers a two year period, and we're talking about momentum traders who use a three-month lookback.  At the end of three months, if the market has trended up, those traders will pile on (by purchasing shares) and thereby push the market further up.  Over the next three months, if the market has trended down, they will pile on (by selling shares) and thereby push the market further down.

Now, as I believe everyone on all sides of this debate has explicitly recognized, this effect requires sufficient presence of momentum trading in the market in order to be meaningful and it can be counteracted by other forces in the market.  But the greater the share of the market that momentum traders make up, the stronger the gyration-amplification effect will be, and the less susceptible it will be to counteraction by other market participants.

While I agree that a DM trader trading with a 6 month lookback period will only make trades that reflect the prior 6 months price momentum of the securities he tracks, I do not agree that there will necessarily be increased market volatility (using your definition of volatility here)  for the coming 6 months because of those trades.  A trade made based on the prior 6 month lookback period can either increase or decrease the subsequent 6 month volatility, depending on what the price movement of the for the security in question in following 6 months turns out to be.

 ie.  If I buy into the S&P because of its superlative performance for the prior past 6 months, and over the subsequent 6 months the S&P goes down, won't I have decreased the S&P's forward 6 month volatility?


brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #922 on: October 26, 2015, 08:02:55 PM »
It's an epidemic, and I've got it bad.

We must have it worse than the others, because we seem to be the only ones still here.

Quote
While I agree that a DM trader trading with a 6 month lookback period will only make trades that reflect the prior 6 months price momentum of the securities he tracks, I do not agree that there will necessarily be increased market volatility (using your definition of volatility here)  for the coming 6 months because of those trades.  A trade made based on the prior 6 month lookback period can either increase or decrease the subsequent 6 month volatility, depending on what the price movement of the for the security in question in following 6 months turns out to be.

ie.  If I buy into the S&P because of its superlative performance for the prior past 6 months, and over the subsequent 6 months the S&P goes down, won't I have decreased the S&P's forward 6 month volatility?

Yes, at the execution of each trade, it will only be the past trend (the one that occurred over the trader's 6 month lookback period) that gets amplified, not the future trend that has, at that point in time, yet to occur.  But 6 months later, the same thing will happen, and again 6 months after that, and so on, so that every market swing that occurs over each of the rolling 6 month periods gets amplified by the momentum trader's trading activity, with the cumulative effect being an increase in semi-annual volatility.

To borrow an analogy from your day job, it's as if someone is watching the output of an old-fashioned EKG machine as it's being drawn and, at regular time intervals, gives the needle a nudge in whichever direction the line trended over each interval.  After several of these intervals have elapsed, the line drawing produced by the EKG will show more volatility (it will have more pronounced spikes and dips).

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #923 on: October 26, 2015, 10:14:57 PM »
To borrow your analogy if at the moment that you push the ECG needle upwards, (because of your measurement of its net prior movement) the needle has already started moving downwards then you will have decreased the velocity of the movement of the needle with your action. You will have decreased volatility.

The effect of your action on the velocity of price movement is completely dependent on the movement of the needle at the moment you interact with it.

And if your concern is a different time period (or distance on the rhythm strip) then your actions effect on volatility is totally dependent on that subsequent movement which you have no influence on.

So the question is really what does intermediate term momentum really predict? Does it predict price movement at the moment that you trade? (I honestly think not.) Does it predict price movement for the subsequent month? (I sure hope so), does it predict price movement for the following 2 months? (No idea.)

Either way it has to enhance that price movement for some periods and detract from it for others. 

So it is very difficult to imagine that you can really know whether or not it increases the volatility of the market as a whole (whatever that means.)

And I would still argue, that the only way  that a trader can effect whether a market chart ends up looking  more like MD2 corp or more like Brooklyn corp is whether or not he increases the velocity of the price movement at the moment he interacts with the market.
« Last Edit: October 26, 2015, 10:19:41 PM by milesdividendmd »

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #924 on: October 27, 2015, 01:05:33 AM »
I can't implement the Global Equity Momentum strategy at the moment because I haven't found a suitable EAFE fund yet.
Of course I don't want to build my own EAFE by buying three or more different ETFs, the trading costs would be too high.

So I need another fund combination to try this out.
I was considering an European index fund (e.g. Eurostoxx 600), MSCI World ex Europe or S&P500, MSCI EM. But I am not quite satisfied with this. Does anyone have any better ideas?

How do I perform backtesting of my desired fund combination? Do I have to find and download the historical index data and then manually compare the numbers?

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #925 on: October 27, 2015, 07:13:08 AM »
To borrow your analogy if at the moment that you push the ECG needle upwards, (because of your measurement of its net prior movement) the needle has already started moving downwards then you will have decreased the velocity of the movement of the needle with your action. You will have decreased volatility.

You will have decreased volatility over the short-term period in which the needle was moving downwards, but if you zoom out in your perspective to the longer-term period that includes the period in which the needle's overall movement was upward, you will have increased volatility.  If you rip off a short piece of the rhythm strip (is that what it's called?) showing only the shorter term period, you will see a downward dip whose amplitude was reduced by your action.  But if you rip off a longer piece showing the longer time period, you will see an overall upward spike (with a small reversal downward near the end) whose amplitude was magnified by your action.

Quote
And if your concern is a different time period (or distance on the rhythm strip) then your actions effect on volatility is totally dependent on that subsequent movement which you have no influence on.

So the question is really what does intermediate term momentum really predict? Does it predict price movement at the moment that you trade? (I honestly think not.) Does it predict price movement for the subsequent month? (I sure hope so), does it predict price movement for the following 2 months? (No idea.)

No, you can affect volatility without knowing the future.  As I've explained (and as Sol originally explained), it is the self-reinforcing amplification of past trends that magnifies volatility, not the amplification of future trends that are yet to be.  If you pile on to every price swing, you amplify them, and over time thereby cause the trajectory of the price chart to look more like MilesDividendMD, Inc. than BrooklynGuy Corp.

Quote
The effect of your action on the velocity of price movement is completely dependent on the movement of the needle at the moment you interact with it.

Neither the needle nor the price of an asset can have a velocity (or momentum or a trend of motion) at a literal instant in time.  This is the essence of the point I've been trying to make.  It is only over a given period of time (which could be minutes or days or years or billionths of a second, but some period of time it must be) that these properties and patterns exist.

When you say what matters is only the movement at the "moment" you nudge the needle/execute the trade (which, again, can't literally mean the snapshot in time when it happens, but rather must mean some exceeding small period of time before the trade is executed approaching--but not reaching--zero), you are just reverting to the use of a shorter-term period for your measurement of volatility.  No matter how short you define that period, I can posit an even-more-micro version of milesdividendmd who similarly protests that no, your so-called effect on "instantaneous" volatility was actually not so, because it was not instantaneous enough, because that micro-version of you is adopting a trillionths-of-a-second perspective instead of the billionths-of-a-second (or whatever) perspective actual-you is using.

So a DM trader's trade does amplify volatility at the zoom-level that is high enough to see that trader's own lookback period, even if the trade goes against the price trend that occurred in the seconds, or minutes, or days leading up to the trade.  And volatility over the months- and year-long time periods impacted by momentum traders using lookback periods in the 3-12 month range (unlike volatility over, say, centennial time periods) does matter to the average stock market investor, and amplifying that volatility therefore can fairly be described as disruptive to the market market or contributory towards market instability.

Of course, once again, this disruptive effect would be meaningful only if there is enough momentum trading to cause it in non-negligible way.  It is a gross understatement to say that the effect of milesdividendmd's own puny trades in the market, like the blowback of a gnat's wings on an EKG needle, is negligible.  But the cumulative effect of sufficient numbers of momentum traders, like sufficient numbers of sheep grazing on the common grounds, is a different story.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #926 on: October 27, 2015, 08:50:12 AM »
Brooklyn,

We are going in circles here, but we are getting closer to an actual understanding I think.  We obviously have to get a bit more Talmudic.

I am not arguing that you can not effect volatility without knowing the future. I am arguing that you can not know whether your action will increase or decrease volatility without knowing the future.

I think the argument will be more productive if we ignore investing for a minute and just discuss our model (the ecg) to keep it simple.

I will make some statements. Tell me which you think are wrong.

In our metaphor "price movement" is the instantaneous velocity of the needle, and "volatility" is the average velocity of the needle over some time period/strip length.

My essential argument has always been that you can only effect the instantaneous velocity of the needle at the moment you interact with it.

You are arguing that you can effect the average velocity of the needle for different time periods regardless of which way the the needle is moving at the moment you effect it (and I agree.)

But I argue that the determination of whether or not you have increased or decreased the average velocity of the needle is completely dependent on the length of the strip you select. If you select a length of strip where the net positional movement of the needle is in the same direction as your push, you will have increased the average velocity, but if you select a different strip length where the net positional movement of the needle is in the opposite direction to your push you will have decreased the needles average velocity.

Same action, same movement trajectory, different effects on average velocity.

And since this average velocity is an unfixed (time period dependent) variable, I am arguing that the only way that it makes sense to measure whether or not a needle pusher increases or decreases the average velocity of the needle is to measure the average effect of his needle pushing actions at the moment they are committed.

(I am purposefully ignoring the scale argument for now, but I'm happy to engage in it later once we have come to some sort of consensus on this essential issue.)
« Last Edit: October 27, 2015, 09:36:21 AM by milesdividendmd »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #927 on: October 27, 2015, 09:35:22 AM »

I can't implement the Global Equity Momentum strategy at the moment because I haven't found a suitable EAFE fund yet.
Of course I don't want to build my own EAFE by buying three or more different ETFs, the trading costs would be too high.

So I need another fund combination to try this out.
I was considering an European index fund (e.g. Eurostoxx 600), MSCI World ex Europe or S&P500, MSCI EM. But I am not quite satisfied with this. Does anyone have any better ideas?

How do I perform backtesting of my desired fund combination? Do I have to find and download the historical index data and then manually compare the numbers?

I think you would do well if you had access to a low-cost S&P 500 fund, Developed Europe fund, and developed Asia fund.

Assuming you do not have to pay for transactions, the cost difference would be negligible. If you do have to pay for transactions, then you will need a lot of money to overcome that price drag on the strategy.

You could implement the strategy one of two ways. Allocate to S&P 500, or developed Europe, or developed Asia, or short term treasuries, depending on the results of your Lookback period.

OrAllocate to S&P 500, or developed Europe PLUS developed Asia (50/50), or short term treasuries, depending on the results of your Lookback period.

In terms of making your measurements at the end of the month, I use the perfcharts function on this site.

Www.Stockcharts.com.

Before implementing my strategy I did a lot of back testing on…

Portfoliovisualizer.com.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #928 on: October 27, 2015, 10:29:14 AM »
but we are getting closer to an actual understanding I think.

Yes, I think so too.  This phase of the discussion is, in my view, a perfect illustration of the type of productive discourse often to be had in this unique corner of the internet.  But, given how we seem to have cleared the room, at this point you and I may be the only ones benefiting from it.

Quote
I will make some statements. Tell me which you think are wrong.

In our metaphor "price movement" is the instantaneous velocity of the needle, and "volatility" is the average velocity of the needle over some time period/strip length.

My essential argument has always been that you can only effect the instantaneous velocity of the needle at the moment you interact with it.

You are arguing that you can effect the average velocity of the needle for different time periods regardless of which way the the needle is moving at the moment you effect it (and I agree.)

But I argue that the determination of whether or not you have increased or decreased the average velocity of the needle is completely dependent on the length of the strip you select. If you select a length of strip where the net positional movement of the needle is in the same direction as your push, you will have increased the average velocity, but if you select a strip length where the net positional movement of the needle is in the opposite direction to your push you will have decreased the needles average velocity.

Same action, same movement trajectory, different effects on average velocity.

I agree with absolutely everything in the quoted text above (subject to the philosophical clarification that, as in physics, the term "instantaneous velocity" cannot, I think, literally be taken as a measure of velocity at a frozen instant in time, but must instead refer to the velocity over an exceedingly small period of time that is close enough to zero to render it beyond the scope of concern), and I think it rearticulates exactly what I've been trying to say.

So isn't it at odds with the following statement from the beginning of your post?

Quote
I am arguing that you can not know whether your action will increase or decrease volatility without knowing the future.

As you just explained, you can know whether you have increased (or decreased) volatility without knowing the future.  And the "volatility" you are increasing (or decreasing) depends on the time frame under consideration.  If the person interfering with the EKG (or ECG?  are they the same thing?) needle is using a lookback period short enough to matter in light of the strip length under consideration, his actions will increase volatility in a way that matters.

Consider this rhythm strip (which no needle-pusher interfered with):



If a needle-pusher had interfered with the needle using a one-second lookback period, his action would have caused the dips and spikes to be more pronounced, correct?  If you look at the same rhythm strip reflecting his actions, wouldn't you see higher volatility in the strip?

If, on the other hand, he used a one-minute lookback period (and his needle push happened to occur in the space of this strip), we have no way of knowing what effect his action would have on the volatility covered by this short strip (but his actions would have a known effect on longer-term volatility over a strip length showing, say, a five minute period (but the repetitive, non-random pattern of this strip probably makes it a bad example for illustrating that point)).

Quote
And since this average velocity is an unfixed (time period dependent) variable, I am arguing that the only way that it makes sense to measure whether or not a needle pusher increases or decreases the average velocity of the needle is to measure the average effect of his needle pushing actions at the moment they are committed.

Yes, but you still have to choose a time period in order to calculate the "average effect" of his needle pushing actions.  And if you choose a time period equal to the lookback period, the needle pushing action will always have the effect of increasing the average effect over that period.  And if the lookback period is short enough for this amplification effect to occur over time periods/strip lengths that actually matter (which would not be the case if the lookback period were, say, 24 hours, if we only care about a one minute strip length), then the needle pushing actions will in turn also matter.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #929 on: October 27, 2015, 12:01:19 PM »

I agree with absolutely everything in the quoted text above (subject to the philosophical clarification that, as in physics, the term "instantaneous velocity" cannot, I think, literally be taken as a measure of velocity at a frozen instant in time, but must instead refer to the velocity over an exceedingly small period of time that is close enough to zero to render it beyond the scope of concern), and I think it rearticulates exactly what I've been trying to say.

Physics tells us that you can have a velocity at a single moment in time, though you will not be able to measure it in a snapshot, but that doesn't really matterfor the purposes of our discussion.

Quote
So isn't it at odds with the following statement from the beginning of your post?


I am arguing that you can not know whether your action will increase or decrease volatility without knowing the future.

No I don't think its at odds at all.  Your effect on subsequent average velocity is completely dependant on the position of the tracing at the point in the future at which you measure the position of the tracing.

Quote
As you just explained, you can know whether you have increased (or decreased) volatility without knowing the future.  And the "volatility" you are increasing (or decreasing) depends on the time frame under consideration.  If the person interfering with the EKG (or ECG?  are they the same thing?) needle is using a lookback period short enough to matter in light of the strip length under consideration, his actions will increase volatility in a way that matters.

A couple points of disagreement here.

1.You can know the whether or not you have increased the velocity of the needle at the moment you interact with it.
2. You can not know how you have effected the average velocity of the needle until you know the position of the needle (and hence the net dorection of movement of the needle) at some point in the future.
3.  The only significance of the lookback period for the purposes of this discussion is that it determines what direction you push the needle.  (This direction will change depending on the length of your lookback period.)

Quote

Consider this rhythm strip (which no needle-pusher interfered with):



If a needle-pusher had interfered with the needle using a one-second lookback period, his action would have caused the dips and spikes to be more pronounced, correct?  If you look at the same rhythm strip reflecting his actions, wouldn't you see higher volatility in the strip?

No!  Either he will increase or decrease the velocity of the needle depending on his lookback period's signal and the direction of the needle's movement at the moment he interacts with it!  And either he will increase or decrease average velocity based on the position he pushes and the net direction of needle movement at the time you decide to measure it. This is important.  Why do you think you  would necessarily see higher "average velocity" of the needle?  what if he pushes against the needle?  What if he pushes in a direction opposite to the net movement of the needle at whatever point you choose to measure it in the future?


Quote
If, on the other hand, he used a one-minute lookback period (and his needle push happened to occur in the space of this strip), we have no way of knowing what effect his action would have on the volatility covered by this short strip (but his actions would have a known effect on longer-term volatility over a strip length showing, say, a five minute period (but the repetitive, non-random pattern of this strip probably makes it a bad example for illustrating that point)).

Again, the only effect that the size of your lookback period has is that it determines which direction you push the needle.  There is no obvious relationship between the size of your lookback period and the direction of movement of the needle at the moment you interact with the needle or the net direction of movement at any time point the future.

Quote
And since this average velocity is an unfixed (time period dependent) variable, I am arguing that the only way that it makes sense to measure whether or not a needle pusher increases or decreases the average velocity of the needle is to measure the average effect of his needle pushing actions at the moment they are committed.

Yes, but you still have to choose a time period in order to calculate the "average effect" of his needle pushing actions.  And if you choose a time period equal to the lookback period, the needle pushing action will always have the effect of increasing the average effect over that period.  And if the lookback period is short enough for this amplification effect to occur over time periods/strip lengths that actually matter (which would not be the case if the lookback period were, say, 24 hours, if we only care about a one minute strip length), then the needle pushing actions will in turn also matter.
[/quote]

Why do you think that "if you choose a time period equal to the lookback period, the needle pushing action will always have the effect of increasing the average effect over that period?"

I see no reason for this statement to be true.  What if you choose a one month lookback which gives you the signal to push up, and one month later the needle is in a position lower than it was at the moment you interacted with it?  Obviously in this scenario, your action would have decreased the average velocity of the needle for that time period.

Please make your case, because I am truly baffled by this claim.  I think this is the essential point of our disagreement.

PS I still suck at formatting quotes.  Sorry!
« Last Edit: October 27, 2015, 12:30:53 PM by milesdividendmd »

brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #930 on: October 27, 2015, 01:44:48 PM »
Quote
In terms of making your measurements at the end of the month, I use the perfcharts function on this site.

Www.Stockcharts.com.

Before implementing my strategy I did a lot of back testing on…

Portfoliovisualizer.com.

Thanks, but unfortunately unusable for me since these sites only show American stocks and funds, in US currency. Which is a big issue, e.g. S&P500 is up "only" about 48% in the last three years in US currency, but +80% in Euros.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #931 on: October 27, 2015, 01:59:37 PM »
Physics tells us that you can have a velocity at a single moment in time

There are philosophical arguments that it doesn't.  Because one of the components of velocity is a rate (a function of time), velocity cannot exist in the absence of time.  Some googling revealed the following claim, which, in substance, is the same as my "philosophical clarification" above:

Quote from: Epimethean Imaginings: Philosophical and Other Meditations on Everyday Light
When mathematicians talk about instantaneous velocity, they are not really talking about movement in no time.  In the case of uniform motion in a straight line, it is simply derived from the measured total distance travelled over a measured period of time.  In speaking of "instantaneous" velocity, mathematicians are not mobilizing a self-contradictory notion that somehow manages to combine being at an instantaneous position with passing through that position at a certain speed.  In the case of variable velocity (as in uniformly or non-uniformly accelerated motion), "instantaneous velocity" is an idea--an unreachable goal--of ever more precise tracking of velocity over smaller and smaller intervals of time to the point at which inaccuracies are unimportant.

I'm not sure whether or not this subtopic matters for purposes of this discussion (it's definitely relevant, but perhaps not dispositive), but either way it is fun to discuss :)

Quote
Why do you think that "if you choose a time period equal to the lookback period, the needle pushing action will always have the effect of increasing the average effect over that period?"

Because the direction of the needle-push will always be in the same direction as the trend over that period (given that the trend itself is what determined which direction to push).

Quote
I see no reason for this statement to be true.  What if you choose a one month lookback which gives you the signal to push up, and one month later the needle is in a position lower than it was at the moment you interacted with it?  Obviously in this scenario, your action would have decreased the average velocity of the needle for that time period.

At the "one month later" point, another push is due.  The signal will tell you to push in whichever direction average velocity took over that month.  So, over the two month period, the "swing" of each of those single months got amplified by your push, meaning that overall volatility over the two-month period got amplified.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #932 on: October 27, 2015, 02:13:09 PM »
Physics tells us that you can have a velocity at a single moment in time


Quote
Why do you think that "if you choose a time period equal to the lookback period, the needle pushing action will always have the effect of increasing the average effect over that period?"


Because the direction of the needle-push will always be in the same direction as the trend over that period (given that the trend itself is what determined which direction to push).


No.  That is simply not true. an example to illustrate why this is so.

 Imagine on Day 0 that your signal is "1," then on day 25 your signal is "100" then on day 30 your signal is "75".  Your signal will then be to push "up"  (since the movement for the lookback period is +75) but at the moment you push up the needle is already moving down.

30 days later the price could be greater than or less than or equal to 75, which will determine whether or not your effect on the average price of that 30 day period was to increase or decrease the price movement.  If it is greater than 75 you will have increased the average price movement, and if it is less than 75 you will have decreased it.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #933 on: October 27, 2015, 02:29:24 PM »
No.  That is simply not true. an example to illustrate why this is so.

 Imagine on Day 0 that your signal is "1," then on day 25 your signal is "100" then on day 30 your signal is "75".  Your signal will then be to push "up"  (since the movement for the lookback period is +75) but at the moment you push up the needle is already moving down.

30 days later the price could be greater than or less than or equal to 75, which will determine whether or not your effect on the average price of that 30 day period was to increase or decrease the price movement.  If it is greater than 75 you will have increased the average price movement, and if it is less than 75 you will have decreased it.

I think you may have misunderstood me.  I was saying that your effect at each trade will always be to amplify the trend that occurred in the lookback period that preceded the trade, not the trend that will occur in the lookback period that will succeed the trade.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #934 on: October 27, 2015, 02:43:37 PM »
No.  That is simply not true. an example to illustrate why this is so.

 Imagine on Day 0 that your signal is "1," then on day 25 your signal is "100" then on day 30 your signal is "75".  Your signal will then be to push "up"  (since the movement for the lookback period is +75) but at the moment you push up the needle is already moving down.

30 days later the price could be greater than or less than or equal to 75, which will determine whether or not your effect on the average price of that 30 day period was to increase or decrease the price movement.  If it is greater than 75 you will have increased the average price movement, and if it is less than 75 you will have decreased it.

Of course your action will be in the same direction as the average price movement for the prior 30 days, but this is a post facto event.  Your prior 30 day period determined the direction of your action, but your action had absolutely nothing to do with the causation of the prior 30 days price movement.

In other words an action cannot cause a change that has already happened.  Actions can effect subsequent events but not previous events

So to make the claim that an action causes a change of any kind, you must by definition be talking about a time period that includes events after the event.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #935 on: October 27, 2015, 03:03:57 PM »
Of course your action will be in the same direction as the average price movement for the prior 30 days, but this is a post facto event.  Your prior 30 day period determined the direction of your action, but your action had absolutely nothing to do with the causation of the prior 30 days price movement.

In other words an action cannot cause a change that has already happened.  Actions can effect subsequent events but not previous events

Of course; I have not been trying to argue that you can magically retroactively affect occurrences in the past through actions in the present.

Quote
So to make the claim that an action causes a change of any kind, you must by definition be talking about a time period that includes events after the event.

Yes.  I am saying the same thing that sol originally said -- that piling on to every price swing will amplify each price swing.

We were getting so close to understanding one another, but there now seems to be another disconnect and I'm not sure what it is.

We've probably long ago lost all spectators to this conversation, but, in the off chance that anyone out there is still reading along and can identify the source of the disconnect that we both seem to be missing, I think it would be helpful to get a fresh pair of eyes looking at this.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #936 on: October 27, 2015, 03:52:13 PM »
This exercise falls apart when we start to use imprecise language like "piling on to every price swing."  Such claims muddy your elegant ECG model.  Let's stay disciplined and stick to the model.

Why don't I make an series of claims.  You can tell me which one you believe to be false.

1.  An action can only cause a change subsequent to the action.
2.  The only way for an agent to effect change directly is through action.
3.  the only action possible to effect the average needle velocity is to push the needle up or down.
4.  Pushing the needle up or down can have no effect on any event that has transpired prior to the push.
5.  The only relevance of the lookback period is that it defines which direction the needle is pushed.
6.  The direction that the needle is pushed does not predict whether the needle will be higher or lower at any later time.
7.  Whether or not pushing the needle increases instantanious needle movement depends only on the direction of the push and the direction of the needle at the moment it is pushed.
8.  Whether or not the needle being pushed increases average needle movement depends only on the direction of the push and the subsequent position of the needle at some point in the future.
9.  The direction of the push can not predict the direction of movement of the needle at the moment it is pushed.
10.  The position of the push can not predict the future position of the needle at any future time.
11.  Because the lookback period can only effect the position of the push, but cannot predict price movement subsequent to the push, there is no predictable positive or negative relationship between the direction of the push and any subsequent velocity or position of the needle.
12.  The period of the lookback period tells you nothing about future needle positions regardless of the time period of the lookback or the time period of the subsequent measurement.
13.  Because the future position of the needle is a variable that is independant of the direction of the push, and the determination of whether or not the average price movement of the needle is dependant only upon the direction of the push and the direction of subsequent price movement, it can not be said that there is any relationship between the lookback period (which effects nothing but the direction of the push,)and the average needle velocity, regardless of the time period of the lookback, or the subsequent point of measurement.

 
« Last Edit: October 27, 2015, 04:02:47 PM by milesdividendmd »

sirdoug007

  • Pencil Stache
  • ****
  • Posts: 585
  • Age: 43
  • Location: Houston, TX
Re: Dual Momentum Investing
« Reply #937 on: October 27, 2015, 04:14:36 PM »
I can't even bring myself to read this stuff anymore. 

Any trade by anyone who types anything on this board is pissing into the ocean.  Let's remember that just the NYSE, which is a subset of all tradable shares, has billions of shares change hands every day! 



High frequency trading is a much bigger issue that DM will ever be. 

In other news, the market has made quite a run recently off the September lows.  Will this mean a whipsaw for the 11/1 6 month lookback DM trade?  It's looking damn close!


milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #938 on: October 27, 2015, 04:23:44 PM »
I can't even bring myself to read this stuff anymore. 

Any trade by anyone who types anything on this board is pissing into the ocean.  Let's remember that just the NYSE, which is a subset of all tradable shares, has billions of shares change hands every day! 



High frequency trading is a much bigger issue that DM will ever be. 

In other news, the market has made quite a run recently off the September lows.  Will this mean a whipsaw for the 11/1 6 month lookback DM trade?  It's looking damn close!

Don't read it!  Brooklyn and I enjoy this stuff.

As of yesterday I would stay in SHY, had about a 1.5 % margin.  We'll see!

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #939 on: October 27, 2015, 06:15:18 PM »
No.  That is simply not true. an example to illustrate why this is so.

 Imagine on Day 0 that your signal is "1," then on day 25 your signal is "100" then on day 30 your signal is "75".  Your signal will then be to push "up"  (since the movement for the lookback period is +75) but at the moment you push up the needle is already moving down.

30 days later the price could be greater than or less than or equal to 75, which will determine whether or not your effect on the average price of that 30 day period was to increase or decrease the price movement.  If it is greater than 75 you will have increased the average price movement, and if it is less than 75 you will have decreased it.

I think you may have misunderstood me.  I was saying that your effect at each trade will always be to amplify the trend that occurred in the lookback period that preceded the trade, not the trend that will occur in the lookback period that will succeed the trade.

I am not sure what you were saying here,  maybe you can redefine what you meant by this.

Quote
I was saying that your effect at each trade will always be to amplify the trend that occurred in the lookback period that preceded the trade, not the trend that will occur in the lookback period that will succeed the trade.

You don't really believe that making a trade can amplify a trend in the past if you agree that it has no effect on past price movements.  or if "volatility" means "average price movement over a period of time" as we had previously agreed, and you also agree that a trade cannot have an effect on past price movement, then you should also agree that a trade can not possibly "amplify" ( or diminish) past volatility, only future volatility.

Final try: if you argue that  a trade increases volatility, by definition the past price trend is wholly irrelevant to this determination, since a trade can never effect past price volatility.

« Last Edit: October 27, 2015, 06:24:26 PM by milesdividendmd »

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Dual Momentum Investing
« Reply #940 on: October 27, 2015, 07:35:31 PM »
13.  Because the future position of the needle is a variable that is independant of the direction of the push, and the determination of whether or not the average price movement of the needle is dependant only upon the direction of the push and the direction of subsequent price movement, it can not be said that there is any relationship between the lookback period (which effects nothing but the direction of the push,)and the average needle velocity, regardless of the time period of the lookback, or the subsequent point of measurement.

I'm thoroughly confused by this one, but if I'm understanding it correctly I disagree with at least two aspects of it.  First, the future position of the needle is not independent of the direction of the push, because the push is a cause whose effect is upward or downward pressure on the needle (so, in the absence of counteracting forces, the needle's future position will be higher or lower as a direct result of the push).  And second, the determination of "average movement" or "average velocity" will depend on the given time period (because that will determine the range of movements or velocities whose average is being determined).

I am not sure what you were saying here,  maybe you can redefine what you meant by this.

Quote
I was saying that your effect at each trade will always be to amplify the trend that occurred in the lookback period that preceded the trade, not the trend that will occur in the lookback period that will succeed the trade.

This was just a restatement of the same observation that was recognized as being "axiomatically correct" earlier in the thread -- the trade will always be in the same direction as the overall trend that occurred during the lookback period.

Quote
You don't really believe that making a trade can amplify a trend in the past if you agree that it has no effect on past price movements.  or if "volatility" means "average price movement over a period of time" as we had previously agreed, and you also agree that a trade cannot have an effect on past price movement, then you should also agree that a trade can not possibly "amplify" ( or diminish) past volatility, only future volatility.

Final try: if you argue that  a trade increases volatility, by definition the past price trend is wholly irrelevant to this determination, since a trade can never effect past price volatility.

An action in the present cannot have an effect on occurrences in the past but it can cause the pattern of the those occurrences to continue.  By trading in the same direction as the overall trend that occurred during the lookback period, the trader "amplifies the trend" not by magically changing history but by contributing to the trend's persistence into the present.

Cathy

  • Handlebar Stache
  • *****
  • Posts: 1044
Re: Dual Momentum Investing
« Reply #941 on: October 27, 2015, 08:15:13 PM »
Physics tells us that you can have a velocity at a single moment in time

There are philosophical arguments that it doesn't.  Because one of the components of velocity is a rate (a function of time), velocity cannot exist in the absence of time. ...

As used in physics, "instantaneous velocity" means something very specific. Here is the formal construction. Suppose that an object moves in 1D space. Its position at time t is given by x(t) where t and x are both real numbers. The object's instantaneous velocity at time t is defined to be the value v that makes the following proposition be true: For every ε > 0, there exists some δ > 0 such that for any rt satisfying |t-r| < δ, it is true that |(x(r)-x(t))/(r-t) - v| < ε. The notation |·| denotes magnitude.

The quantity given by this definition turns out to be very useful in physics. To the extent I understand the philosophical objection, it appears to be that this quantity is not really a "velocity". In other words, the complaint is purely about the name given to the quantity rather than about the utility of the quantity. However, it is a ubiquitous use of the word, and it is not especially misleading because this definition reduces to the discrete concept of velocity when the inputs are discrete.
« Last Edit: October 27, 2015, 08:37:06 PM by Cathy »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Dual Momentum Investing
« Reply #942 on: October 27, 2015, 09:08:26 PM »
Quote
13.  Because the future position of the needle is a variable that is independant of the direction of the push, and the determination of whether or not the average price movement of the needle is dependant only upon the direction of the push and the direction of subsequent price movement, it can not be said that there is any relationship between the lookback period (which effects nothing but the direction of the push,)and the average needle velocity, regardless of the time period of the lookback, or the subsequent point of measurement.


I'm thoroughly confused by this one, but if I'm understanding it correctly I disagree with at least two aspects of it.  First, the future position of the needle is not independent of the direction of the push, because the push is a cause whose effect is upward or downward pressure on the needle (so, in the absence of counteracting forces, the needle's future position will be higher or lower as a direct result of the push).  And second, the determination of "average movement" or "average velocity" will depend on the given time period (because that will determine the range of movements or velocities whose average is being determined).

Actually the push only effects needle movement at the moment force is exerted. So the future movement (ie the movement that occurs after the finger has lost contact with the needle) is independent of the push. But that's not actually so important. What is important is that future movement or positions of the needle are not predicted by the direction of the push.




Quote

I am not sure what you were saying here,  maybe you can redefine what you meant by this.

Quote
I was saying that your effect at each trade will always be to amplify the trend that occurred in the lookback period that preceded the trade, not the trend that will occur in the lookback period that will succeed the trade.

This was just a restatement of the same observation that was recognized as being "axiomatically correct" earlier in the thread -- the trade will always be in the same direction as the overall trend that occurred during the lookback period.


An action in the present cannot have an effect on occurrences in the past but it can cause the pattern of the those occurrences to continue.  By trading in the same direction as the overall trend that occurred during the lookback period, the trader "amplifies the trend" not by magically changing history but by contributing to the trend's persistence into the present.

I understand what you are saying here. But the disconnect is that what has occurred before (and after I would argue) the finger pushing the needle is wholly independent of the pushes momentary effect on needle velocity.

In other words in retrospect you can say that there is a single right answer to the question of whether or not the finger has increased or decreased the needles velocity at the moment of the push, but you can never say whether the finger has increased or decreased the needles average velocity definitively, because that answer is wholly dependent on the random choice of where you decide to start measuring and where you stop measuring the needles movement.

And while the trend of the lookback period can always predict what the direction of the push will be, it can never tell you what the direction of the needle movement will be at the moment you push on it.
« Last Edit: October 27, 2015, 11:43:19 PM by milesdividendmd »

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #943 on: October 27, 2015, 09:39:30 PM »

Physics tells us that you can have a velocity at a single moment in time

There are philosophical arguments that it doesn't.  Because one of the components of velocity is a rate (a function of time), velocity cannot exist in the absence of time. ...

As used in physics, "instantaneous velocity" means something very specific. Here is the formal construction. Suppose that an object moves in 1D space. Its position at time t is given by x(t) where t and x are both real numbers. The object's instantaneous velocity at time t is defined to be the value v that makes the following proposition be true: For every ε > 0, there exists some δ > 0 such that for any rt satisfying |t-r| < δ, it is true that |(x(r)-x(t))/(r-t) - v| < ε. The notation |·| denotes magnitude.

The quantity given by this definition turns out to be very useful in physics. To the extent I understand the philosophical objection, it appears to be that this quantity is not really a "velocity". In other words, the complaint is purely about the name given to the quantity rather than about the utility of the quantity. However, it is a ubiquitous use of the word, and it is not especially misleading because this definition reduces to the discrete concept of velocity when the inputs are discrete.

Beautiful Cathy. I wish I had the math skills to understand what you wrote.

What I do recall is that velocity is not a relative property.

I remember asking my professor why relativity tells us that time moves slower for the guy on the train when the guy on the ground is moving at the exact same speed relative to the guy on the train, and he pointed out that only the guy on the train had accelerated.

milesdividendmd

  • Handlebar Stache
  • *****
  • Posts: 1913
  • Location: Portlandia
    • Miles Dividend MD
Re: Dual Momentum Investing
« Reply #944 on: October 27, 2015, 10:26:07 PM »

Quote
In terms of making your measurements at the end of the month, I use the perfcharts function on this site.

Www.Stockcharts.com.

Before implementing my strategy I did a lot of back testing on…

Portfoliovisualizer.com.

Thanks, but unfortunately unusable for me since these sites only show American stocks and funds, in US currency. Which is a big issue, e.g. S&P500 is up "only" about 48% in the last three years in US currency, but +80% in Euros.

Of course. Great point. I'm of limited use to you.

Is there a yahoo finance or equivalent in your country?  That should work for backtesting.

Also here is a nice article for you to digest.

http://www.dualmomentum.net/2015/06/dual-momentum-for-non-us-investors.html?m=1




brainfart

  • Stubble
  • **
  • Posts: 182
Re: Dual Momentum Investing
« Reply #945 on: October 28, 2015, 12:32:05 AM »
> Is there a yahoo finance or equivalent in your country?  That should work for backtesting.

Yes, of course. But I'm not exactly sure how to do this backtesting. Will investigate, thanks.

> Also here is a nice article for you to digest.

Thanks again, saw that one months go, read it and didn't fully understand all its details, and then forgot about it. And of course, unfortunately, it deals with all the English-language currencies and the Yen as a worst case example, but not the Euro. Guess that currency it's not that important considering only a few hundred million people use it ... ;)
Will read it again and see if it makes more sense now.

So end of month approaching. What is the momentum saying?

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Dual Momentum Investing
« Reply #946 on: October 28, 2015, 01:43:14 AM »
MOD NOTE: miles has been temporarily banned for a few weeks for behavior in other (unrelated) threads. He will be back in a few weeks, and, I'm sure, be happy to continue any ongoing debates or answer any questions left here in the meantime.

Just a courtesy message for those involved in this thread.  :)

Please PM me or any other mod with concerns.  Cheers!
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.

frugalnacho

  • Walrus Stache
  • *******
  • Posts: 5055
  • Age: 41
  • Location: Metro Detroit
Re: Dual Momentum Investing
« Reply #947 on: October 28, 2015, 10:13:47 AM »

Actually the push only effects needle movement at the moment force is exerted. So the future movement (ie the movement that occurs after the finger has lost contact with the needle) is independent of the push. But that's not actually so important. What is important is that future movement or positions of the needle are not predicted by the direction of the push.

Wait a minute, isn't this the momentum thread? I was under the impression that several people reading the momentum of the needle and pushing it up will not only cause the needle to go up (the immediate effect while force is being exerted), but that will in turn make prices higher, which will in turn trigger more fingers pushing the needle up*, which will in turn make prices higher, etc. creating a cycle.  The future price is not independent of the push because it carries that momentum forward.

*Not necessarily from DMers, but from everyone, because this is fundamental to how momentum, and this strategy, work in the first place.

frugalnacho

  • Walrus Stache
  • *******
  • Posts: 5055
  • Age: 41
  • Location: Metro Detroit
Re: Dual Momentum Investing
« Reply #948 on: October 28, 2015, 10:19:18 AM »
If you or I trade our entire portfolio at any one time it effects the global capital markets much as a tear drop effects the salinity of the ocean.

We have no measurable market impact by virtue of our small size.

No one is arguing that any individual does.  No individual rain drop feels responsible for the flood.  And it's damn near impossible to measure the magnitude with which one rain drop did contribute to the flood.  But I still know as a matter of logic that in aggregate those rain drops did indeed cause the flood.

K-ice

  • Pencil Stache
  • ****
  • Posts: 982
  • Location: Canada
Re: Dual Momentum Investing
« Reply #949 on: October 30, 2015, 09:15:00 AM »
Sorry if this has been addressed. I have skimmed through much of this thread and it is hard to pull out information.

So I have a simple question to those using DMI.

With how much of your net worth are you actually doing DMI?

I have heard it is best with tax advantaged accounts.  Also, some of us have a large amount of net worth in real-estate (prime residence & rentals).

To answer my own question.

Out of my investable assets (cash, stocks, bonds etc.)  I would consider DMI with about 50%.  Out of my total net worth it would be less than 20%.


Some of you may have a number figure like, I will try DMI with $10K $100K etc.

I am just trying to get a feel for what people are actually doing.

Thanks in advance for sharing.



 

Wow, a phone plan for fifteen bucks!