Some signs of the right TCs - page 29
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
Optimise many custom characters - MultiTester.
Thanks, I'll try to investigate. Will it work on linux (I'm migrating to winndows, but I won't complete this process any time soon)?
Thanks, I'll give it a go. Will it work on linux (I'm migrating to Windows, but I won't complete this process any time soon)?
WinAPI is used there. Perhaps this will help you to answer your question.
I may be wrong, but if we need to test on a sufficiently large number of custom characters, then the only option is to make a single very long custom character out of them. If we need optimization on a sufficiently large number of custom characters, then this method doesn't seem to be suitable anymore.
I can vaguely see some possibility to apply "generalized correctness" in building a portfolio of one EA with different parameters. For this purpose we conduct optimization on the set of transformed quotes. The transformation of quotes, for example, consists in cutting them into a given number of pieces and then rearranging and gluing them in all possible orders.
The identity of transformations in TS is a logical requirement for the same operation on an inverse and multiple instrument. I don't understand the idea. why make up a long symbol with different characteristics on different chunks and break up optimizedparameters on these chunks by time, and why optimization on a large number of symbols is not suitable. Price analysis should give some characteristics of the analyzed chunk and on the basis of the characteristics choose optimization parameters, without prior analysis the same optimization on different behavior of the number series will give different quality optimization results.
I would add the main thing:
Utopian fantasies
However I've been exploiting exactly such a system for several years. Here it is, it's been trading on demo for 2.5 days. During first 2 days it has made 104k from 10k, during last half a day it already started getting close to 2 m. In total it is 175 times in 2.5 days.
Utopian fantasy
But I have been using this exact system for several years. I've been trading on demo for 2.5 days already. During first 2 days it has made 104 thousand from 10 thousand, during last half a day it already started to get close to 2 million. In total it is 175 times for 2.5 days.
Yes, of course, everything is necessarily in the past, or will be in the future. And in the present deals are the longest 10 sec.
We already had something similar here, there was a show from fxsaber with pumping money from one account to another in one brokerage house. We have already gone through something similar in the past. There was a show at fxsaber with money transfer from one account to another in one brokerage company.
***
And I have another immodest question - do you have nothing better to do? What is the sense and joy in testing systems with secnuded trades on a demo account?
It's a bit of a departure from the subject. We can put another question, what can and can't be used in TC to make it work equally on the reverse and multiple symbols. And how and when it may be useful. In the conditions, the input is three parameters per tick, the prices Bid, Ask and Time. We trade a relative change in prices, subject to the mandatory loss of the trade, i.e. we open and close at prices opposite to the direction of the trade. And this is thehanalysis, i.e. there are no signals from the outside. In pure mathematics the condition of identity gives a necessary condition, but the logic of two prices or spread per trade makes it difficult. In general the logic of less, more equals, as well as timing with accuracy to the tick with the analysis of the order direction and at the same time all parameters in relative units. What improves. Uniformity of work with different price behavior. What makes it worse. Profit against a mathematically incorrect TS.
Why?
Why?
The same result of analysis and optimisation, moving away from random fitting. Analysis and optimisation in a mathematically correct TS will give more stable results.
it's the logic of two prices or spreads per trade that makes it difficult.
bid/ask must be algorithmically symmetrical.
bid/ask must be algorithmically symmetric.
Yes, this condition must be met, it makes it difficult, but it does not make it unsolvable. It also leads to a split calculation algorithm, and does not make it possible to calculate via mathematical formulas. At least I haven't encountered that in the formula it is possible to take into account a sign depending on any parameters of a series and direction of the transaction.