Price action waves, retracements, and trends. Do you think AI will make a dent in these foundational patterns in the next 5 years?
Interesting possibilities. When Precter and Frost did their work for their EW book, the markets were dominated by retail investors... who feel emotions that AI doesn't/won't. Granvilled called the 3 up-legs of bull markets, Disbelief, Belief, and Overbelief... each psychological. And then there is the interventionist aspect of the Fed and the PPT in the markets these days.... will all that turn EW on its head??
I wonder about this too. I think they will make markets more predictable, I could be wrong. They will all program similar algo's further solidifying market patterns. Markets will do the same things over and over again. Just my 2 cents, I hope my very basic answer is correct. I'm sure others will have more to add.
I feel that the markets are definitely more "whippy" or choppy over the last couple of years. I think AI will change the face of trading as more lower/medium traders/trading companies fall off the cliff...of course there are always more to fill in the vacancies but the odds are very slim but these will die off at a much faster rate than previous waves of newbies . I think the big boys will be battling each other more and we will see more faster slamming up and down. Maybe its my paranoia writing this...lol.. I think basic structure is still there but widening of stops is required unless one has found the "grail" and size is just enough that one can hide in the grass and not be easily detected.
ai is as silly as the people who believe it can be applied to a dynamic market. ai is a close relative of curve fitting it can not learn, it can only curve fit the past.
It already has. Doesn't necessarily lead to price stability given that most algos employed are very near sighted.
a branch of AI (supervised ML) has a bigger danger of curve fitting v. unsupervised ML. But in the hands of a skilled trader, (ie Human-in-the-loop)+SML can be a "force multiplier"
I don't believe it's possible to have a thread like this without participants having written in complete ignorance of 1987. Here: https://duckduckgo.com/?q=crash+of+'87+program+trading&t=ffsb&ia=web Y'all take a deep, freakin' breath. D'ya realize that 2018 was the first year of historically-normal volatility in 5? years? More? C'monnnnnnn.
It's hard to tell, but last year many quant hedge funds failed and quants got blamed for simply not being good at trading. So this will continue happening, while AI strategies may be undermining each other. While if any approach would be better than all others than it itself would change the rules of the game and fail against itself. And when the rules get changed, there will be new failures, AI will get blamed and taken out of the game, while we'll be back where we started.