Mar

23

The market's repertoire of rhythms extends past human grasp. Sometimes it seems to make no sense at all, at least to me.

Sometimes, things seem to become clear. Just as in Afro-Cuban music, a strong voice - the "mother drum" in bata - dominates the counter rhythms of the smaller drums, sometimes the Fed's announcements dominate the backdrop of lesser voices — Chinese monetary authorities, fixed-systems followers, and what have you.

Earnings season has a peculiar rhythm. But it's ever-changing, based on which companies are strongest at the time.

One quality the market shares with music, good music, anyway, is "always the surprise." Bach, Mozart, Beethoven were all masters of deception and expert at weaving rhythms across bars. Beethoven's sforzandi, unexpected sharp accents, and sudden pianissimos, will be appreciated by all traders.

Back in the '90s, when I was the editor for the stock coverage, a humorous bond reporter at Bloomberg had a saying when stocks had yet another amazing jump: "Stocks ONLY GO UP," he would say, rolling his eyes knowingly, meaning just the opposite. No good musician plays loud all the time.

Victor writes: 

I am thinking of ways to quantify the rhythms of markets. Instead of looking at what others do, critiquing it, and then augmenting, I thought I'd just take a crack at thinking of it my own way.

Music rhythms would seem to be a good starting point. The rhythms that kids are taught are those they can step or clap or slap to. They can be fast or slow to start with. And I would look to see if the number of moves in a minute is fast or slow and how this changes. The slapping would involve moves from separate markets occurring in the same time period. When we step, the first step is the accented one and that's a good way to look at moves within a period. Is it the first step that's always the biggest, and what happens when the second or third step in a period is the biggest?

I would look next at the rhythms of big moves. They obviously are reversing now, with some big Tuesdays: February 27, -58; March 8, +22; March 13, -28; and March 20, +8. Naturally this kind of stuff isn't predictive in general or else it would come out in the standard time series programs. But on occasion, it comes back and forth to an inordinate degree and the question becomes how to find it.

Animals often migrate at the same time of year to the same places even when transported geographically. One wonders if the migrations of markets after big moves have a fixed place in the price firmament that they go back to. Or is it just in time, like the conventional seasonal stuff that one can expect from the migration? Last year, prices went way down in May and migrated back the last part of the year. This year the migration started in February. The month ended with the three old bags ("a woman her age would never show her posterior to a camera") acting in concert with the rhythmic release of the perennially bearish message from the Sage.

The rhythms of political announcements always seem to follow a circular path. They start with a loose cannon doing something that hits into something else. Then others join the act. One typical sequence involves worry about inflation, based of course on a preview of an upcoming release, then the release of the number, then the big bond fund guy saying he's bearish, then the perma-bears finding other inflationary things, then the opportunistic movement in certain nations that benefit from this or that energy price, and finally the rhythm ending with the release of the next number, or the quieting influence of an open market meeting.


Hoaglin has some great diagrams of rhythms in the body. And the body has different rhythms that it responds to as molecules bounce into each other and create disturbances throughout other more complex molecules, thus upsetting the usual homeostatic methods. One market makes a big move, perhaps silver, and it spills over into others in a rhythmic sequence, perhaps an up in energy, and then a decline in stocks. It's not over until the initiating market has its move back down as was the actual case with the recent bloodbath and recovery, which seemed to have the elements of rhythm of all the ones I mentioned.

Of course, the rhythms have to be combined with the melodies. The speed of the moves has to be counted with the steps between those moves, sometimes big and sometimes small. And I like the way they quantify melodies in the Joy of Music and in the statistical studies of music intervals that have so much resonance with markets.

A more humdrum approach to rhythms, which I take, is to look at the rhythms of patterns. How often do the 3-day moves with their eight possible directions: —, –+, -+-, -++, +++, ++-, +-+, + — repeat? Is it a first order Markov process vis-a-vis these eight patterns, and what is the correlation between the closeness of each of the last three moves to these three patterns, and future moves? I recently ran some rhythm stuff with open, open to close, and open, and found some ministerial randomness with t's all below one, but enough evidence of non-randomness to get me thinking about rhythms on the whole.

I know enough about rhythms to know that they feel like the basic rhythms come from within the body, like the beating of the heart, and they can model it with rhythms based on the mathematics of African rhythms. Whatever quantifications they are making in bringing African rhymes and Latin rhythms into the heart beat problem would seem to be a natural for extension into the market.

I am fortunate to know someone with perfect rhythm and she is the coeditor of this column and I am going to ask her how she would try to trade in the market if she knew nothing else but markets. Perhaps other musicians with perfect rhythm might have similar expert opinions as to where market moves would be going based on their knowledge and oneness with rhythms in markets. Certainly these experts would be more prone to give good calls than the eminent people who have passed the tests of the mystical societies of America that are licensed to forecast the market.

The market's open now, and I haven't read any of the dozens of books I have on rhythms lately, but after I do and study it on the Net, perhaps I'll have some other ideas. For sure, my colleagues will be able to augment my preliminary fast ideas on this and guide others and me in proper directions.

George Zachar comments:

Perhaps other musicians with perfect rhythm might have similar expert opinions as to where market moves would be going based on their knowledge and oneness with rhythms in markets. Certainly these experts would be more prone to give good calls…

An interesting way to test this would be to submit representations of various tradeables in various time increments to musical prodigies who are naive about markets. I am thinking particularly of junior and senior high school students, who could have sufficient musical training and experience, without having been exposed to what passes for financial and economic wisdom in the popular press. 

Ken Smith writes:

In harmony with Victor's piece on music, rhythm, I attempted to write a melody with three notes. I am having difficulty conveying this little ditty because the note symbols for music are not available in email text messages.

I've tried before to get symbols to end up as they were written when they appear after I've sent them. Somewhere in the Internet circle symbols sent in email get warped, become hijra. Meanings are thus distorted.

So maybe someone can help here. The musical symbols for this simple melody would be symbols for the Dollar, Mark, and Yen, just three notes.

Create a melody using these notes - they are real notes, after all. Then choreograph a dance for the melody. Add lyrics. Create permutations and program computers to trade dollar, mark, yen - according to the melody.

"A salient feature of markets is temptation." (Syncreticus)

Todd Tracy writes:

Everyday I am inspired by the list and become more humble. In the business of music I had done well being rather sure of myself. That confidence came about from having practiced hours daily for 20 years. And even then I had much to learn. Afro-Cuban percussion was one of those things I knew nothing about until the day that my roommate brought home four percussionists. I didn't know at the time that they would be living and practicing in my living room for two years. And yes, they had many percussionist friends. The neighbors didn't seem to mind. They played all day, ten drummers strong, and then went on to their gigs at night.

One guy, Jacques, studied African rhythms. His guru was Babinga. Another guy, Blake, studied Cuban fusion. His guru was Giovanni Hidalgo. Davey was into Indian drums, Egyptian bells, and all sorts of experimental world music. Josh was a well-rounded guy who did it all. Their friends were mostly jazz funk kit players.

At any rate, I was doing 80 hours a week at the record company but on occasion they would let me sit in with them during rehearsals. When it came to the Congolese and Senegalese rhythms I had to learn to play the pattern given to me and not concentrate on the patterns the other guys were playing. The African stuff doesn't resolve like western music. Each part is simple; the complexity comes from the layering. Euro rhythms resolve every measure. Four beats to a measure at tempos ranging from 60-130 beats per minute. The African stuff would resolve many measures out, like ten equivalent western measures. It felt as though it was random until, with incredible anticipation, the resolution was at hand.

The Latin stuff was different in that the Cubans, Haitians, and Puerto Ricans had fused the African rhythms with western melodies. The most important part to the rhythm was the clave (wooden sticks that ring out when struck). The clave would be a simplified version of the rhythm. Then came the congas. They would play a rhythm called a Tumbao. Again, you had to concentrate on your part but synchronicity was achieved and resolved after just a couple of measures.

I was completely humbled by all that I did not know. But soon, through repetition, I found I had a whole new arsenal. These guys would play until their hands bled every day as they developed the incredible muscle memory needed to counter western rhythmic intuition.

Now the straight up rap beats are simple in that they are looped (kind of like rock music). But it is the anticipation of that resolution that concerns us with the market rhythms. In hip-hop the kick is on the one and the three; the snare is on the Two and the Four. The snares are played late to increase the anticipation. This lateness is the most important part, in fact, so important that rap artists actually consider the two and the four as the one and the three.

All of the rhythms resolve. There are problems in programming the beats in that there is a finite number of places to put each beat within a measure (460 ticks per beat) and the velocity of each beat is set at a value 1-127. We can, however, increase the resolution by doubling the BPM and by fine-tuning these anticipations and resolutions. I am studying the Quantlet Booklets so that I could one day break down the market rhythms as is being shown to me by the list members through the graciousness of Victor and Laurel's benevolence.

As far as what I think the S&P index will do from a musician's perspective is resolving to 1450 after channeling a bit more.

Laurence Glazier writes:

It is very tempting to apply my knowledge of music in selecting trades, though I like to follow grounded mathematical principles at this stage. I would note that much of what we consider the theory of music was derived by the posthumous analysis of the works of the one and only JS Bach (the Moses of music?), which like much technical analysis is seductive but not necessarily predictive. I work on the principle that part of this analysis represents laws of musical reality empirically testable, but not in the normal way. As Leschetitsky said, "Where words end, music begins."

Of the technical analysts of music, Schenker is particularly interesting, while those who have enjoyed "The Glen Miller Story" may have observed the appearance of another significant analyst, Schillinger.

Having said that, I believe the analogies with market rhythms, while not necessarily predictive, would be very valuable as part of a real-time virtual reality program reflecting the current state of play in the markets, and pose the question whether users of such a system would do better if they were more musical.

Victor Niederhoffer adds:

There is something rhythmic in the moves of bonds and stocks together, over and above the comparative rates of return that the Duo and Dodger have quantified. And it's like the monkey rope that Melville describes, where when one goes down and the other has to follow. But there is much thrashing around as the turbulence from the whales temporarily overrides the inextricable bond.

And in that context the bonds, after setting a 19-day low at 11,202, are still up 2/3 of a point or about 1/2% on the year. And the stocks, after setting a 19-day high at 1445, are up about 1/2% on the year. Regardless of that it's what I used to call an ugly day and the rhythm is very bad for both when a big decline in one occurs in conjunction with a big rise in the other. Something has to give, and as Berlioz would say in reviewing Beethoven, you know it's going to return.

George Criparocos writes:

The two days preceding the big note (02.27, the resonant, memorable one) had the bonds making a rhythmic intro analogous to what is expected when the largest instrument of all, the bass, announces a change in melody.

Since then, the contrabass, cellos, and violas (10s, 5s, 2s) are keeping the resonance, while the bass returns. The clarinet (Yen) is hanging around its 200MA set like a rope, refusing to let go of the anticipation and the piano (stocks) are all over the pentagram, in 1/16th intervals: four days low, four days high, four days flat, four days high.

The rhythm seems to be analogous to a symphony, lets say in F major. The allegro is in progress and I anticipate that the andante should follow in a molto mosso way.

James Sogi adds:

Todd's analysis of African rhythms resolving over eight or 12 bars or multiples rather than the simplistic four beat 16 bar square "rock" structure is right on the beat.

One of the most basic rhythms popular in the blues is called the shuffle. It is a short-long, short-long, short-long, similar to the heartbeat or train on the track, da-dum, da-dum, da-dum. This basic rhythm underlies many more complex patterns.

Applied to the market after a small beat, there the long bar, the "shuffle." The count often does not capture the rhythm, just as European musical notation does not carry information relative to rhythm. That is an odd omission. A shuffle might be notated as straight quarter notes, but played as doted quarter and eighth note sequence and designated as a shuffle, all the musicians know right away what it means.

The rhythm can get behind the pocket, giving a laid back feeling, like the end of last week. Or the rhythm can get ahead of the beat, like disco, like last month's drop.

The middle of the pocket of the beat is the march's oom-pah, oom-pah, even beats. The rhythms will swing from behind the "pocket" and give the music different feels. This is very difficult to quantify because the interaction of the multiple players is complex and the "feel" is a subtle thing to capture. Musicians know this.

To capture this in the market is a difficult matter. The main difficulty is the time structure. A structure stretched out over weeks is difficult to feel for human rhythmic sense as our rhythm is based on the heart and walking, and resides in the feet and heart and head motions. So it's hard to feel the market rhythm without condensing the time and looking at the numbers or speeding it up on a replay as an interesting exercise.

Russ Sears writes:

To Be With Me
by Russ Sears

Chic chic ca dee!
The Bluebird on our clothes line sings to me.
Come home, come home,
To be, to be,
 to be with me.

Kar Reeee! Kar Reeee!
The Bluejay mocks the hawk in perfect key
Go! Clear! Go! Clear!
Not free, not free,
No meal is free!

Tit tit ra lee!
The glorious Lark boost for all to see
Stay back, Stay back,
Match me, match me
You cant match me.

From Vincent Andres:

I am thinking of ways to "quantify" the rhythms of markets.

I didn't test it yet (will probably do so sooner or later) but the already known track of Hurst/Hölder/ exponents seem to me to be a possibly good piece of measurement.

Another possible tool could be wavelets.

Also, I recently came across a paper melting wavelets + Hölder curves : L'analyse par ondelettes, in Science, Vol.119 Sept. 1987. Yves Meyer, S. Jaffard, Olivier Rioul. The paper is in French. Very certainly progress have been made since this paper was published.

 


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search