Trading Catechism

Rieger-Nishimura.svg.png
 
Is positive edge just negative edge time-reversed? It seems reasonable and plausible, and yet, it is not obvious to me at all.

Many people will state that it is easy to find a good trading system. Just find a really bad one (of which there seem to be an infinite number of them) and reverse its signals. But does this work in the real world?

Take a look at the graphic two posts above. It says you can get from false to truth by negating your way there. That is standard mathematical logic. But I claim that market logic does and does not allow (respectively) the two axioms that make this possible:

  • The law of excluded middle. Markets allow a fuzzy "in between",
  • double negation. Markets do not allow this because you have many in between truths in between false and true.
The reason almost certainly has to do with Entropy. Still, a part of me wonders what can make it work, for then I could be twice as profitable.
 
Last edited:
noddyboy said:
How does this apply to trading?
More...
Models can either be something you cooked up, or with these tools, you can build a model using standard techniques from AI and machine learning. In essence, this is a bottom up approach to data, but with the ease of a useful interface (by interface I don't mean GUI, I mean linguistic interface) where human beings can think about the problem domain at a high level succinctly instead of the language or the mathematics.

This greatly facilitates model. So you are using the computer for what it is good for, and in theory you can sandwich yourself in between the data and the concepts.

These particular tools atomically support stochastic + probability models, making them potentially very useful. I am particularly interested in Markov logic models.
 
nitro said:
Models can either be something you cooked up, or with these tools, you can build a model using standard techniques from AI and machine learning. In essence, this is a bottom up approach to data, but with the ease of a useful interface (by interface I don't mean GUI, I mean linguistic interface) where human beings can think about the problem domain at a high level succinctly instead of the language or the mathematics.

This greatly facilitates model. So you are using the computer for what it is good for, and in theory you can sandwich yourself in between the data and the concepts.

These particular tools atomically support stochastic + probability models, making them potentially very useful. I am particularly interested in Markov logic models.
More...

Makes sense. I am feeling that in 10 years robots are smarter than humans, and wouldn't even need us to write code. This will all be obsolete.
 
noddyboy said:
Makes sense. I am feeling that in 10 years robots are smarter than humans, and wouldn't even need us to write code. This will all be obsolete.
More...
Ten years might be a bit of a stretch. But in fifty? We will all be hamsters that just get in the way.

The future of human race is probably some sort of hybrid cybernetic beings.
 
You must log in or register to reply here.
Back
Top