Optimise an EA and get the best of the optimised ones. - page 35
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
Aleksey Vyazmikin:
It's OK. It is accounted for. Coded to the league.
37 regcodes.
At the moment, left to be reoptimised:
Betting myself:
1 NZDCHF EMAFlatSAR
Test period 22.04.17 - 22.04.18, forward from 22.09.17
Launched
I think a triple flight advisor would be good.
I remember such a magazine from my childhood.
If you have become fixated on a profitable trend, part of the profit should be invested in opening mini trades against the trend.
on the flat, while it is oscillating in search of a trend, instead of losing by stop losses, let it open and close at the highs and lows of the channel, with small profits, and when the deal went into a drawdown, it has already opened a deal on the trend.
this is the approach, take the system as a whole. just don't let it turn out to be a hedged wheel.
I think each trader can choose the subtleties of each part for himself.
Aleksey Vyazmikin:
All good.
38 rgcodes.
I think a triple flight advisor would be good.
I remember such a magazine from my childhood.
If you have become fixated on a profitable trend, part of the profit should be invested in opening mini trades against the trend.
on the flat, while it is oscillating in search of a trend, instead of losing by stop losses, let it open and close at the highs and lows of the channel, with small profits, and when the deal went into a drawdown, it has already opened a deal on the trend.
this is the approach, take the system as a whole. just so long as it does not turn out to be a hedged wheel.
NO.
I think one of the biggest mistakes is to take TS in its entirety, as a whole.
TC is a set of techniques, each of which brings something different. Moreover, the most important techniques are those of accompaniment. Therefore I use four of them in my League, while the current trend is determined by only two variants, and entries are taken in two variants.
Besides, I've already said - as I made sure a long time ago, complex TP, that takes into account a bunch of factors, as you suggest (let's speak about "you"), affects the same way the simplest TP does. That begs the question - why do we need a "triple flight EA" when it is so much easier to get three separate simple EAs?
All games with MM like martingale, locking, grids - all this is evil, they work if the system is working. If the system does not work - they can only save it in theory. The simplest calculation shows that the use of martin really works, but the profit is extremely small, and not worth the cost. If we set reasonable trade values, the number of martingale-series and low probability of loss - the deposit becomes huge with a very small profit (but correct).
Well...
All TCs that have failed are over-optimised...
Plugging in the next symbol. CADJPY
Seven ЕМАш ТС are already present on it.
Now we need:
EMAFlatDTS
And all channel:
ChnTrendDTS
ChnTrendSAR
ChnTrendSP
ChnFlatSP
ChnFlatSAR
ChnFlatRTS
ChnTrendRTS
ChnFlatDTS
Running EMAFlatDTS for over-optimisationNO.
I think one of the grossest mistakes is to take TC as a whole, as a whole.
TC is a collection of techniques, each of which brings something different to the table. Moreover, the most important techniques are those of accompaniment. Therefore I use four of them in my League, while the current trend is determined by only two variants, and entries are taken in two variants.
Besides, I've already said - as I made sure a long time ago, complex TP, that takes into account a bunch of factors, as you suggest (let's speak about "you"), affects the same way the simplest TP does. This begs the question - why do we need a "triple flight EA" when it is much easier to get three separate simple EAs?
I think I'll report in a month. All games with MM like martingale, locking, grids - all this is evil. If the system does not work - they can only save it in theory. The simplest calculation shows that the use of martin really works, but the profit is extremely small, and not worth the cost. If we set reasonable trade values, the number of martingale-series and low probability of loss - the deposit becomes huge with a very small profit (but correct).
The difference between using different EAs, allows you to run on history all of the systems you run together and see the results.
I'm going to try to trade both Flat and Trend without Loks against the Trend. I think I'll report my results in a month.
The difference of using different EAs, allows you to run all the systems you run on history together and see the results.
Exactly, in my opinion, it is a mistake to consider a "mix" of TS.
Yes, it improves the characteristics of the result (expected payoff is added, but the sum of standard deviations becomes the root of the square of the sum of squares of deviations, which is less than the sum).
But, it is necessary to investigate TC only separately - to see "weak points".
dimanep:
i'll try to trade both flat and trand, without locks, against the trand. i think i'll report it in a month or so.
Well, well.
My trading robot has 8 trend and 8 flat systems working on each symbol. I don't need to wait a month. But what's the use of it, I don't understand. If the TS (doesn't) work - then it (doesn't) works independently of the others. The sense in the joint use is only in the case of a significant relationship between the TS - either when using data of each other, or through correlation of pairs. But it is better to avoid using correlated TSs, for correlated pairs arbitrage TSs should be used.