Do I trust the strategy tester? - page 3
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
Judging by the local grail-grabbers, having a grail does not add much money for some reason )
So, the phrase can be shortened to:
You can't expect to make any extra profit.
Andrey, since when here about it? I hope all is well with you?
They work the same way, not because they have a similar algorithm, but because they have the same author.
There is no group effect because there is no group of authors, the brain is the same.
Yes, yes, the local oligarchs have hundreds of percentages every day. But apart from you, I don't see any such.
And about "the same" - it's not, and that's exactly the problem, there's no stability.
There is a very simple way to check whether a tester can be trusted.
You run your robot for 2019, choose the best parameters and run them for 2020.
You look at the result. I think those who have been here for a long time already know exactly what you will see :)
If the robot is made by a local grailmaker, the check is a bit more complicated to get around some ingenious moves.
The robot is written by me. The first one. I am still in training, I am far away from "local grailmakers". That is why the question was about trust in the tester's results. Maybe I should try more years (not only 2019, 2020)?
The robot was written by me. The first one. I'm still learning, I'm a long way from the "local grailmakers". That's why the question was about confidence in the tester results. Maybe by more years (not only 2019, 2020)?
Well one of the purposes of the tester is to pick the parameters that will be the best on the tested range, and it does a great job with that. The problem is that it offers no guarantee that the parameters will continue to work for any length of time. It seems logical that the larger the tested range, the more robust the algorithm is to new, previously unknown quotes. But this is false logic. Most likely, the larger the range to be tested, the greater the number of false (as well as true) signals will be discarded. In general, the more range you capture, the more entropy the processed information will have. And you cannot make money on chaotic data due to its unpredictability.
So, it's better to trade on more predictable things like stocks. Technical and fundamental analysis will help to find more or less right entry point, and then you can forget about them for a long period of time (after all, there is no swap robbery - hold as long as you want).
Well one of the purposes of the tester is to pick the parameters that will be the best on the range being tested, and it does a fine job of this. The problem is that this gives no guarantee that the parameters will continue to work for any length of time.
Build the logic chain a little further...
If there is no guarantee that the chosen parameters will work further, then the issue BEFORE is not with the tester, but with your strategy and the settings that were obtained in the process of FITTING to HISTORY...
If the strategy is based on the correct IDEA, laid down during development, then this strategy will work profitably for quite a long time...
In general, TESTING is almost the CONCLUSIVE stage in strategy development...
Start with the RIGHT IDEA...
Build the logical chain a little further...
If there is no guarantee that the chosen parameters will work further, then the issue BEFORE is not in the tester, but in your strategy and in the settings that were obtained in the process of FITTING to the HISTORY...
If the strategy is based on the correct IDEA, laid down during development, then this strategy will work profitably for quite a long time...
In general, TESTING is almost the CONCLUSIVE stage in strategy development...
Start with the RIGHT IDEA...
If you have a working idea, you don't need a tester. Hence the conclusion - a strategy tester is a useless thing :)
The trouble is that 99% here are exactly fitting the parameters to the story, moreover, then trying to sell it in the marketplace.
If you have a working idea, you don't need a tester. Hence the conclusion - a strategy tester is a useless thing :)
The problem is that 99% of people here are adjusting the parameters to the story, and moreover, trying to sell it on the market afterwards.
If you have a working idea, you don't need a tester. Hence, a strategy tester is a useless thing :)
A working idea and a profitable strategy are not comparable things...
Just TESTING makes sure that your naughty hands have not messed up in this long process when moving from idea to strategy.