Machine learning in trading: theory, models, practice and algo-trading - page 1694
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
interesting article, or rather the second half of it... it's nothing just to read if you have nothing to do
https://proglib.io/p/trade-learning
So your best solution is to do nothing... That's funny...
You have to do a lot of things, a lot more than it seems. But it is important to rely on the right foundation, the basis of which is relevant data, which by the way does not need to buy for millions, but you have to collect from year to year, then come the right chips, this is art, analytics, you need to understand how the market works and what big (informed) investors react to, you generally need to clearly understand how the global economy works, chips from price indicators is nonsense. Almost no one knows how to prepare targets correctly, it's an art too, but then you also need to trade...
In short, not a ploughed field, terra incognita for most. What they mostly do here is child's play, which has no resemblance to the real algotrading.
For example, I would be happy if someone would tell me how to decrease equity, for example, for a tester.
Above I gave one such "harmful" advice, not to use optimization (GA, grid) of Rice Management together with "search for a pattern" in the form of optimization of indices parameters. Firstly, the optimization of indices is a futile idea, but even if it were something sensible, it would be much easier to geta "grail" with pure maneuvering, fitting lots in TP and SL grid, optimization quickly finds March-like solutions, which are often "grail-like" on the forward, but it does not mean anything, there is no advantage over the market, such "strategies" are also found in SB.
Again, there are a lot of ways to redistribute the frequency and magnitude of profit/losses so that losses become so rare that they simply do not fall into the test, this is the basis of 99% of all grails, take a story an order of magnitude more and everything usually falls into place.
Conclusion: If the strategy does not work on one lot, it is rubbish.Above I gave one such "harmful" advice, not to use optimization (GA, grid) of Rice Management together with "search for a pattern" in the form of optimization of parameters of indices. Firstly, the optimization of indices is a futile idea, but even if it were something sensible, it would be much easier to get a "grail" with pure maneuvering, fitting lots in TP and SL grid, optimization quickly finds March-like solutions, which are often "grail-like" on the forward, but it does not mean anything, there is no advantage over the market, such "strategies" are just like that on SB.
Once again, there are plenty of ways to redistribute the frequency and magnitude of profit/losses so that losses become so rare that they simply do not fall into the test, this is the basis of 99% of all grails, take a story an order of magnitude more and everything usually falls into place.
Conclusion: If the strategy does not work on one lot - crap.Kesha, are trend-following strategies welcome?
Trendy when a trend is probable, and flat when a flat is probable.
In fact, everything is worse with trend strategies, because no one from the majors is regularly pouring into the market, now they make such a "rush" that it is impossible to understand where exactly the majors plan to go, it is better to make money with reverse TP and leave at the first sign of strong movement.Trendy when a trend is likely, and flat when a flat is likely.
They do not invest in the market regularly any more, now they do such "mess" that it is impossible to understand where exactly the big guy has decided to go.However...
I'm running this strategy now.
So far, I'm not happy with the profitability.
i'm sitting here thinking...
I'm sitting here thinking...
In theory, a flat is the same trend, but without the slope.
You need to do a lot, a lot more than it seems. But it is important to rely on the right foundation, the basis of which is relevant data, which by the way you do not need to buy for millions, but you have to collect from year to year, then come the right chips, this is art, analytics, you need to understand how the market works and what big (informed) investors react to, here in general you need to clearly understand how the global economy works, chips from price indicators is nonsense. Almost no one knows how to prepare targets correctly, it's an art too, but then you also need to trade...
In short, not a ploughed field, terra incognita for most. What they mostly do here is child's play which has no resemblance to real algotrading.
It works for me somehow.
"While the smart man was going, the fool crossed the river.
The new version now does not just show "confidence" and direction, but gives an entry point. Five minutes ago I got a buy signal and by the time I'm writing this, the bit has already risen by 150. Everything is public.
It works for me somehow.
"While the smart man was going, the fool crossed the river".
If you are good at trading your forecasts, you better avoid publicity, profits like silence, it's a fact, and if someone yells that he earns a lot or hasa "grail", then surely something is wrong, or a crook (seller of trash), or just a lucky guy who will sell out in a month.
Algotrading is very hard work, not cheap by all criteria, but it is honeycombed with easy profits.
If you are good at trading your forecasts, you better avoid publicity, profits like silence, it's a fact, and if someone yells that he earns a lot, or he has a "grail", then surely something is wrong, or a crook (seller of fake), or just a lucky guy who will sell out in a month.
Algotrading is very hard work, not cheap by any criteria, but pamazan honey easy profits.
I agree, I used to think so myself. But, as you know, in trading it's not the trader who earns, but those who provide services. During gold rush more and more stable earnings were made not by miners, but by those who sold them shovels.
And at the expense of "may jinx", so pray and all will pass))
As you know, it is not the trader who earns in trading, but those who supply services. In the gold rush, it was not the miners who earned more and more steadily, but those who sold them shovels.
Yes, it's a sensible way, a bird in the hand, so to speak, don't bother with cranes, it's troublesome))) But IMHO it's easier to get a job than to rob a very small simpleton, selling them a sham "shovels", we both know that the real shovels would never sell))).
But I'm not judging, in principle I even respect fraud, but the big ones, not petty.