Fr7Participant
Master
Hello Leon, It is a long time since you have written about neural networks again ….. Have you advanced or discovered something better new?
Has anyone got astrategy file working with leos indicator?
LeoParticipant
Veteran
Hi all,
long time without being here connected.
I was astonished with the idea that a program write his own rules of trading. Then I start learning python and more specific Machine Learning, like Random Forest or Artificial Neural Network, then I start to write the program for connect to the API of IG for trading with python, and so on and so on,
Now, after few year of learning and with 1000 euros now I happily buy my house, I bought the Tesla I dreamed and ensure the financial security for my parents and parents in law….
Of course not…
After many research by myself and other people I meet in my journey, the conclusion is: You will never ever predict the future of an stochastic phenomenal like financial markets….
Everyone telling you the opposite is a storyteller.
the conclusion is: You will never ever predict the future of an stochastic phenomenal like financial markets….
Indeed, AI will not help either, but it is so much fun! 🙂
LeoParticipant
Veteran
Yeah, is very fun and I learn new skills in data science indeed.
I also discover with podcast, blogs, reports, papers, etc that there are only 3 ways of winning money in this business:
- Arbitrage: you need a very expensive software and a team of Phd , programers etc. and a lot of capital. I met with someone who works in company and they move 100 millions euros in the first minute of the opening market in order to win several hundred euros… risk free (most of the time), but they microwave connection directly with Frankfurt, etc, etc, etc….
- Fees: you are “hedge fund manager” and even if you get negative Alpha then they pay you commissions, marker makers, spreads, coaching, programming robots for others, YouTuber, etc
- Insiders… which is somehow illegal
Anything else is speculation i.e. gambling…
What I won is a new profession as data science and I do not give up yet in financial markets… but almost.
@Leo, I didn’t try yet your Neural Netwroks programme, but I wanted to congratulate you (1) for going that deep as an individual (as opposed to a company with a team) in a such complex and innovative topic and (2) for being so generous with all of us by sharing your hard work. The most important is the learning journey, not necessarily the result. The problem is sometimes you’re right, but not at the right time. I hope PRT will take this topic very seriously before the competition do so that we can have the option to include (or not) self embedded optimization into our Algos, AI, and the like.
May be a suggestion of improvement is to include invalidation scenarios, i.e. after the Neural Netwroks determines that this setup has been successful more than x on 5min TF, then enter a trade en 5 min TF, then if you have a reverse say in 15 min TF, you exit the trade before you hit your SL.
I wish you a bright future!
Hi Leo,
I note the age of this thread and therefore it may be unlikely you read this but I just want to address a couple of points you have made in order not to discourage others who may just be embarking on a trading journey and stumble across this thread.
After many research by myself and other people I meet in my journey, the conclusion is: You will never ever predict the future of an stochastic phenomenal like financial markets….
Everyone telling you the opposite is a storyteller.
You are correct in your assertion but wrong in your conclusion. You will never predict the future of a stochastic phenomenon like the financial markets but that is never the intention or aim of someone who is consistently profitable. That is the aim of someone who lacks experience and is still honing this craft. The aim of consistently profitable traders is not to predict the future but to analyse past events and use whatever means at their disposal to recognise a similar contemporaneous pattern and put themselves into a position where they are taking a value position against people they are trading against. The aim of this game is not to predict the future but to take value, i.e you have a better chance of an event occuring than the broker is offering. The easiest way to do this is with Options as you can quickly work out what the broker is offering in way of odds but looking at the delta and converting this into the chance you believe that event has. This way you are guaranteed over the course of infinity to make money. You won’t always be right but overtime you will be right more often than you are wrong. Your indicator and neural network could prove invaluable at spotting trends appearing earlier than a human can alert the user to value before the market makers have time to adjust.
Anything else is speculation i.e. gambling…
It is my friend, but speculation, gambling, trading, call it what you will, they are all prima facie the same thing. That doesn’t mean that you can’t make money consistently though, I make around 4% ROI per month on Betfair exchange alone, this is just a betting exchange for sports betting. I don’t arb or have inside knowledge like you said, I only take value time and time again when I know the odds aren’t correct and the price of an event occurring is incorrect. Again as stated above this is no different to trading, and again trading Options offers the best exposure.
Anyway, I shan’t ramble on any further, I just wanted to make sure people aren’t too discouraged, it took me the best part of 15 years to achieve consistent profits in this game, the journey is a long one, but very achievable.
