Fibonacci Queen | Trading Reviews and Vendors - futures trading strategies, market news, trading charts and platforms

Trading Reviews and Vendors

Discuss and review vendors of commercial trading products, trading rooms and services, trading indicators or third-party paid add-ons


Fibonacci Queen

  #46 (permalink)

south africa
Posts: 170 since Dec 2018

Fat Tails View Post
If markets were completely non-stationary and non-ergodic, nobody could possibly trade them. Did ever come to your mind that mass behavior might impose temporary patterns? Or what you recommend to shut down this forum, because discussing trading concepts is a waste of time anyhow?

For deeper understanding of how this works, I would recommend the reading of Robert Axelrods "The Evolution of Cooperation". This book - which is not about markets - has much contributed to my understanding of how markets work.

So fib levels work because there are so many people using them?
Come on. The amount of money being bet using fib levels is absolutely nothing in 2019 as % of all capital bet.

You dodge the question though as far as distributions of what. You said "I have run some probability distributions on fibs. They give you a moderate edge."

So you are saying then that fib levels have a property that a set of random levels do not. I was thinking about this after I posted the response driving around and I don't think this is testable. What are we even saying fib lines do exactly to start with to sample from and have a distribution of what? Is price attracted to them? Is price repelled by them? Does price bounce off them like a bouncing ball hitting a concrete floor? Then we would have to test whatever that is against random sets of retracement lines.

Then we are saying this works the same on monthly real estate prices in Chile, Corn futures data from 1985-1990 AND tick data of the cryptocurrency TRON? Every market displays this because of human psychology and crowd behavior no matter how different the market?

Nonstationary and non ergodic doesn't mean 100% random and unpredictable. A time series doesn't have to be stationary in order to trade it.

Actually, if markets were always stationary and ergodic we would not be able to trade because pricing then would be a simple statistical process. It is not a problem for trading, it is a problem though when talking about comparing a fib level to a random level variable and comparing two distributions across all time and all markets.

I will check out that book, it sounds interesting. To me "Sapiens" by Yuval Noah Harari explains much of what we are talking about here.
We cooperate in large numbers because of our ability to invent fictional narratives. Fib lines are just random lines with a nice backstory.

To me it is one of the most obvious examples of being fooled by randomness.

Reply With Quote