What to feed to the input of the neural network? Your ideas... - page 40
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
1. .... Maybe we should give examples of FFs in the article that take into account not only indicators describing the financial result, but also other indicators affecting it that are implied but not named?
2. I have already written in the first paragraph, and here I will only repeat that people are trying to understand why they need it, and come to understand the alternative to the in-house optimiser with its genetics. What other parameters that are far from the market should be set in FF - it is not clear to most people.
1. Here is an article in which the author describes the work with FF, the article considers as many as 4 examples covered in articles by other authors and gives additionally his own (however, the title of the article emphasises not what is most valuable in it).
2. See para. 2. alternatives are sought, probably because they don't understand the point, they didn't get a normal result - ah, the tester is bad!
There are not many articles on the topic of FF, but they are on mql5.com, recommendations are given, practically useful aspects of optimisation are covered. But I will say this: no one will ever give their universal FF, it is equivalent to revealing the secret of the Grail.
1. Here is an article in which the author describes the work with FF, the article considers as many as 4 examples covered in articles of other authors and gives additionally his own (however, the title of the article emphasises not what is the most valuable in it).
2. See para. 2. they are looking for an alternative, probably because they don't understand the essence, they don't get a normal result - ah, the tester is bad!
There are not many articles on the topic of FF, but they are on mql5.com, recommendations are given, practically useful aspects of optimisation are covered. But I will say this: no one will ever give their universal FF, it is equal to revealing the secret of the Grail.
1. In the article only work with financial indicators, including balance.
2. So far, I come to the conclusion that I can't explain to you what is the reason of scepticism of other participants towards such approaches of model creation. Although, I tried very hard, I suggested to talk about data structure and how to account for it in FF.
See, you again think that others don't understand the topic at all, that's one of the reasons why some people are overreacting. I don't mean that in a reproachful way, it's just feedback.
Yes, I use a rather complex (many criteria) FF, but not for optimisation, but for evaluating the quality of the strategy in the Expert Advisor sold in the market. It uses a completely different approach, which I have not seen described - a preliminary independent study of different sets of parameters with their estimation is performed, and then such sets are randomised with the selected parameters (not just one variant is selected, but also a set). This approach allows to cover a large area of parameters and leaves quite a large variability. Does it work - well, here are two signals on this method works for me - traded a decent amount of trades, while under this MM was not initially optimised.
Don't you realise that with this message you are not extinguishing the conflict but adding oil to the fire?
If you did not do it deliberately, then I suggest everyone pretend that point 8. in Aleksey Vyazmikin 's post simply does not exist.
With this post I emphasised that one should try to understand others before making assumptions about them (and their thoughts). Without understanding, there can be no peace and harmony.
I'm not interested in conflict, just frustrated by the parties' lack of understanding of each other. If there is no understanding, then the forces go to conflict instead of clarifying and challenging each other's issues.
1. The article only deals with financial indicators, including balance sheet.
2. So far, I come to the conclusion that I cannot explain to you the reason for the scepticism of other participants towards such approaches of model creation. Although, I tried very hard, I suggested to talk about data structure and how it can be taken into account in FF.
See, you again think that others don't know anything about this topic, that's one of the reasons why some people are overreacting. I don't mean that in a reproachful way, it's just feedback.
3. Yes, I use a rather complex (many criteria) FF, but not for optimisation, but to evaluate the quality of the strategy in the Expert Advisor that is sold in the market. It uses a completely different approach, which I have not seen described - a preliminary independent study of different sets of parameters with their estimation is performed, and then such sets are randomised with the selected parameters (not just one variant is selected, but also a set). This approach allows to cover a large area of parameters and leaves quite a large variability. Does it work - well, here are two signals on this method works for me - traded a decent amount of trades, while under this MM was not initially optimised.
1. Not only.
2. I don't understand, why talk about it at all? Just speak for yourself.
3. very good. But by using FF you are already doing optimisation, look at the definition of "optimisation". I wish you success and further improvement of your methods, write more articles, more views from different angles is always good.
1. Not only.
2. I don't see why we need to talk about this at all? Just speak for yourself.
3. very good. But by using FF you are already doing optimisation, look up the definition of "optimisation". I wish you success and further improvement of your methods, write more articles, more views from different angles is always good.
1. What for example?
2. To distinguish between the perception of aggression to the individual and to scientific research. You can work with the second one, leading explanations, clarifications. But, you need to understand, and I couldn't explain.
3. Thank you. And I wish you creative success!
1. Like what?
2. To distinguish between the perception of aggression towards a person and towards scientific research. You can work with the second one, leading explanations, clarifications. But you have to understand, and I couldn't explain.
3. Thank you. I wish you creative success!
1. The way to apply FF.
2. Whose aggression? Either be blunt or don't talk about it at all.
3. Okay, reciprocated.
Let's imagine that we already have a lot of predictors that we can't wait to feed to the input of the NS. But, the computer may overload from the excess of incoming data - we don't have millions of dollars for supercomputers. What to do in this case, what optimisation algorithm will be ideal for the task of selecting the most useful predictors, what kind of FF can be invented, and will it be more efficient than standard methods from the economic point of view? This is a question that never ceases to worry the poor :))))) What are your thoughts?
Red should describe green in such a way that one has to look for a global maximum utility. And, what exactly are the "standard" to compare with?
Tried to feed in:
-closing prices
- difference of closing prices of N candles in a row
- difference of closing prices of N candles in a row from all allied pairs on both the euro and dollar sides, on the eurodollar pair
- ratios of candles' shadows to their bodies of N candles in a row
- ratio of closing price to all OHLC of all N candles in a row
- all of the above with N and N*k gaps between candles
And....I've run out of ideas))))
You can get a tick chart and feed the neural network with the number of large (in terms of points) ticks, i.e. somewhere the price jumped by 12 points in an instant, somewhere by 47 .... etc.
Thanks for the idea.
Unfortunately, it is difficult to work with ticks, I work with opening prices.