Machine learning in trading: theory, models, practice and algo-trading - page 2421
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
You are absolutely right, I am very far from Maxim! My conscience doesn't allow me to deceive people!
If you've read Maxim's articles and his statements here, you must have understood his approach - he has many times said that models do not work long and the method itself, which allows you to quickly retrain and use a new model, is important. So maybe you should ask him to retrain his model and compile the EA and post the new version on the Market.
If we want to explain globally why the Expert Advisors fail, the reason is that models on decision trees are built on training samples that are not representative, i.e. the patterns identified on them are found to a lesser extent in the future. Trees are built in parallel or sequentially - their drawback is that they want to describe the entire sample, resulting in very weak patterns which are given high weight in the list, for example 5 observations with a positive outcome was in the training sample of all, and in the future this pattern has ceased to recur. And now imagine that we have not one tree, but an ensemble of trees that contains tens of thousands of leaves that affect decision making, I think it is easy to understand here that the number of combinations of response of these leaves is very large and on training and test samples they simply will not meet in full, and this would not be a problem if the training sample reflected all patterns that any new sample with data after training would.
The bottom line is that an approach that can use consistent patterns over a long time interval to build trees will work in MO. How to identify these patterns is not an easy question, but it is solvable, but even once solved, no one guarantees that these patterns will not change over time.
And in any case, every Expert Advisor has periods when it is not effective due to the fact that the patterns it exploits are implemented to a lesser extent than before. This may be due to a new factor which was not present before or was present for a short period of time, for example, there may be years when the instrument has been in a very sharp trend for almost a year with rare departures into different price ranges and in this case trend experts will lose money and nothing can be done globally.
We should sue! This is outrageous!
Write in the lawsuit that you believed the man, read the articles, dreamed of passive income. And he deceived you.
It is a pity they do not put him on a stake or at least cut off his hand.
We will support you. We are also victims, we trusted him, we crap a lot of code, even a decent indicator came out of a neural network and still no passive income.
That's a good one, but it's not. If cheating is a norm in your life and you don't react to it in any way, then I just feel sorry for you.
I'm an artist, that's how I see it, you can't be offended by that
In your case it would be more correct to say "I am a swindler and storyteller, I like to tell tales and draw pictures".
It's a good one, but it's not. If cheating is the norm in your life and you don't react to it, then I feel sorry for you.
Don't you find it funny? You bought a grail, and it wasn't a grail.
You bought not food, not clothes, not a washing machine, and a bot which is sure to earn because all the previous 20 bots bought earned and all their friends earn, and here OBAMANUUUUUUUUUUUUUUUUUUUUUUU!
In your case it would be more correct to say "I am a fraud and a storyteller, I like to tell tales and draw pictures".
Why did you miss it?
Isn't it funny to you? You bought a grail, and it turned out not to be a grail.
You bought not food, not clothes, not a washing machine, and a bot that just had to earn because all the previous 20 bought bots earned and all my friends earn, and here OBAMANUUUUUUUUUUUUUUUUUU!
Not funny at all! If the product is of poor quality, not checked on forward tests, don't offer it for sale, don't write colorful texts and tell tales.
Your text you are trying to say that the deception of the buyer - it's common, like everyone does, so what, such is the world.
I am trying to tell you that you need to treat the work responsibly, to the buyer with respect, if there was a failure of the system, to notify in a timely manner about it, to stop the work and make adjustments.
That's what professionalism is and command respect! And your world where everyone cheats each other is miserable and boring. Earned money on the principles of deception will give you nothing in life, any person who has gone through this experience will tell you that.
It may seem that way from the outside, everyone has a different view. It would be interesting to see the opposite reaction when the situation changes.
If you do not have a clear view on this, you are the expert who pours non-stop, every trade is loss-making, and you are discussing it quietly.
The respectful attitude towards your customers would be to inform them that they all need to stop trading until they make a change.
But you do not do that, because it is obvious even to you that the algorithm is broken. Why? Because you don't care.
With your text you are trying to say that cheating the buyer is a common thing, like everybody does it, so what, that's the world.
You're trying to pull something that I didn't write and didn't mean. I just gave an assessment of your transaction and I'm not making excuses for anyone.