Do you see patterns in Random Walks?

[SrRuthenate]

Quote from jem:

By traditional definition random would make profit due to skill impossible.

There is quite a difference between random, normally distributed, and the actual fat tailed distributed chaotic market.

Although the math isn't complex nor the principles, the implementation is. With their differences is like if I say E=mc2, now build me a nuke to fry some Japs. Or here is the equation for a rocket XYZ, fly me to the moon and back.

 

[stepan7]

Quote from jem:


I was a profitable trader about 7 years... til they change the rules.

Who are "they" and what are "rules"?

Thanks.

 

[jem]

Quote from stepan7:

Who are "they" and what are "rules"?

Thanks.

Mostly the NYSE.

Our big edge also started to disappear as more derivatives came on the market, more ecns and smaller pricing increments.... allowing discrepancies to be arbed back into line in a more automated manner.

We human traders were replaced by automated traders. The idea that that was random... to me used to be a joke... but I now realize was just savy marketing.

 

[jem]

Quote from SrRuthenate:

There is quite a difference between random, normally distributed, and the actual fat tailed distributed chaotic market.

Although the math isn't complex nor the principles, the implementation is. With their differences is like if I say E=mc2, now build me a nuke to fry some Japs. Or here is the equation for a rocket XYZ, fly me to the moon and back.

Its as difficult as you wish to make it.
What turns out to profitable is finding patterns before others do.
How many instruments to you wish to arb? How much money can you spend on data. How fast can you execute... all allow you to work on smaller and smaller patterns - perhaps across more and more instruments.

But in the end its all the same thing... recognizing a pattern before your competitor.

 

[Samsara]

Quote from jem:

I wrote a much more detailed explanation but I will explain why this thread died out.


The esoteric knowledge needed to profit from random markets is the the knowledge to find profitable patterns and the marketing savvy to change the change the nomenclature and the traditional definition of random.

I generally feel the same way (and also agreed with your pre-edit), but tend to bracket this reading of human nature with my own ignorance. I personally think that much of the math that works to locate certain patterns requires only simple algebra, some calculus for certain applications like measuring rate of change or modeling turning points, and a tool set of descriptive and inferential statistics.

I just feel that the case has not been made for randomness. It seems to me the first principles upon which the EMH is based requires a reified foundation (and those principles don't make intuitive sense to me anyway), and my personal observation -- both seeing simple consistent behavior on one hand, and blown up quant models that were esoterically marketed on the other -- is enough for me to take the same cynical road as you.

That doesn't necessarily mean that there's shit out there I don't understand that works more consistently and is better grounded in its theoretical underpinnings, like modeling flocking behavior in an n-dimensional space of instruments, basing it on a certain picture of mass psychology. I think Maestro was giving a decent go at grounding the fundamentals of his own model.

 

[Hansel H]

Quote from HATEtheRisk:

Samsara sucks.

..if only because it doesn't make sense. If you're a bad person and come back as a chipmunk how can you develop good or bad chipmunk karma in preparation for your next incarnation?

 

[jem]

Quote from Samsara:

I generally feel the same way (and also agreed with your pre-edit), but tend to bracket this reading of human nature with my own ignorance. I personally think that much of the math that works to locate certain patterns requires only simple algebra, some calculus for certain applications like measuring rate of change or modeling turning points, and a tool set of descriptive and inferential statistics.

I just feel that the case has not been made for randomness. It seems to me the first principles upon which the EMH is based requires a reified foundation (and those principles don't make intuitive sense to me anyway), and my personal observation -- both seeing simple consistent behavior on one hand, and blown up quant models that were esoterically marketed on the other -- is enough for me to take the same cynical road as you.

That doesn't necessarily mean that there's shit out there I don't understand that works more consistently and is better grounded in its theoretical underpinnings, like modeling flocking behavior in an n-dimensional space of instruments, basing it on a certain picture of mass psychology. I think Maestro was giving a decent go at grounding the fundamentals of his own model.

I agree... and I think your last paragraph is key, for me. I give the person credit for having the structure to profit from research. But what is flocking really...

I simplify for illustration..

Is it not... I see a pattern in a shorter time period or in a set of markets or data and I can make a profit off that observation because at least for now that data is leading the set of trades I will take to profitablity?

In the 90s I spent a good deal of time looking for data which could lead the stocks and futures I traded. Every once in a while I would find something which worked for a few months at a time. I called it a leading indicator and I knew it would someday cease being a leading indicator.

 

[Samsara]

Quote from jem:

I agree... and I think your last paragraph is key to me. I give the person credit for have the structure to profit from research. But what is flocking really...

I simplify for illustration..

Is it not... I see a pattern in a shorter time period or in a set of markets or data and I can make a profit off that observation because at least for now that data is leading the set of trades I will take to profitablity?

In the 90s I spent a good deal of time looking for data which could lead the stocks and futures I traded. Every once in a while I would find something which worked for a few months at a time. I called it a leading indicator and I knew it would someday cease being a leading indicator.

Makes sense to me to see it that way.

And same here: my constant anxiety is always fixated on the decay of what I currently have. Aside from the intellectual stimulation of looking for and testing hypotheses, in the back of my mind I'm always concerned with differentiating drawdown from edge decay.

I think this might be why randomness is such a perversely hot topic among traders: those who propound the EMH make that anxiety into cognitive dissonance -- that it's a fundamental truth that anything currently working is just a mirage. And being deceived is worse than a fear of change, because it means you're forever building on sand.

 

[intradaybill]

Quote from Antisyzygy:

I go to school part time for my PhD.

No serious university out there accepts part-time PhD students, AFAIK. Please provide a link to your university program.

 

[Mike805]

Quote from intradaybill:

No serious university out there accepts part-time PhD students, AFAIK. Please provide a link to your university program.

Bill,

Even though you have some understanding beyond a basic undergraduate math curriculum, this does not give you the right to question the practices of many "serious" universities.

Making blanket generalizations about PhD's, academics and finance professionals only demonstrates how little knowledge of these topics you possess. Avoid such generalizations in the future if you wish to be taken seriously.

Mike

P.S. It is widely known that lesser educated people tend to question the abilities of those with higher education than themselves, if for no other reason than to combat deep seeded feelings of insecurity and inadequacy. How's that for a blanket generalization?