Login / Register
Add a review
Log in
Sign UP

Excelsior multi-character Expert Advisor implements a popular variant of a recurrent neural network and trades on daily charts. The EA is ready for fully automatic trading on 8 symbols: AUDUSD, EURUSD, GBPCHF, GBPUSD, NZDUSD, USDCAD, USDCHF and USDJPY.

The main difference of this Expert Advisor is that it uses 2 independent recurrent neural networks to calculate the buy and sell signals. Two neural networks compete in such a way that their signals can exclude each other. Since the resulting signal from the two neural networks may change the next day, the adviser may decide to change or close the transaction during its lifetime.


Run the Expert Advisor on one chart. Regardless of your current schedule, the Excelsior Expert Advisor always works simultaneously with 8 symbols.

The EA calculates the values for the daily timeframe as needed, so your timeframe does not matter.

You can close trades manually if you deem it necessary. You can open other trades on the same account or use other indicators and Expert Advisors. This does not affect the work of the Excelsior Advisor.

You can stop the work of the Expert Advisor or restart it, the Expert Advisor resumes its calculations as usual.

[spoiler title=”

The Excelsior ADVISOR runs on a VPS server.

The Expert Advisor closes trades or sets a stop loss along the opening line according to its algorithm when the trend changes. The initial stop loss is indicative and is needed for internal calculations.

Comment – comment.

Fontsize – font size. Change the value if the text “gets out” of the panel borders.

Suffix – suffix. For example, the suffix _i for the symbol EURUSD_i. If there is no suffix, leave the field empty.

Close losing trades (days) – closing of unprofitable trades (the number of days after which the trade has not become profitable). You need at least 2 trades so that their total current profit is positive. It turns out that unprofitable trades will be closed with simultaneous compensation due to profitable trades.

Same signal max – the maximum number of transactions for the same signal. A signal on one symbol is considered to be the same if it is within the range of the same signal. Possible values: 1 – 5. Values greater than 3 create an unreasonably high figure

Same signal range – the range of the same signal. Possible values: 0 – 400. A value of 0 means that there is no identical signal, and the EA opens all possible trades; the value of the maximum number of trades on the same signal is ignored.

Stoploss – stoploss.

Volume – the volume of one trade. Set a specific value (example: 0.01) or a percentage of available funds (example: 0.02%). The Expert Advisor prints a specific volume value on the “Experts” tab when starting or restarting. Always use a specific value for cent accounts or accounts denominated in Bitcoin, rubles, yuan, etc. (that is, not in US dollars).

All default settings (to reset to the default settings, click the Reset button in the context menu of the Adviser properties).

Minimum deposit: 250 USD.

It may be that at the first launch, the Expert Advisor does not have time to download all the data at once. If there are no deals for more than two days, try opening the daily chart of each symbol and scroll it to the left. Then restart the Expert Advisor.

The Expert Advisor uses neural networks, and therefore transactions may differ for different brokers and different types of accounts. Long-term trading results for 3-4 months are similar to the recovery factor, Sharpe ratio, drawdown and other parameters. Trading always corresponds to the history on one account.

The Expert Advisor is ready to work with the default settings. To achieve optimal profitability for your deposit size and leverage, optimize these properties of the Expert Advisor for the last 1 year:

Close losing trades (days): 0 – 60 (step 20)

Same signal max: 1 – 5 (step 1)

Same signal range: 0 – 80 (step 20)

Choose a combination of properties that shows optimal profit and drawdown. See the optimization drawings.


Add review
Reviews ( 0 ):