分配
| 交易品种 | 交易 | Sell | Buy | |
|---|---|---|---|---|
| USDJPY | 15 | |||
| EURJPY | 10 | |||
| GBPJPY | 9 | |||
| GBPUSD | 6 | |||
| AUDJPY | 5 | |||
|
5
10
15
20
|
5
10
15
20
|
5
10
15
20
|
| 交易品种 | 毛利, USD | 损失, USD | 利润, USD | |
|---|---|---|---|---|
| USDJPY | 11 | |||
| EURJPY | 4 | |||
| GBPJPY | -15 | |||
| GBPUSD | -6 | |||
| AUDJPY | 2 | |||
|
10
20
30
40
50
60
|
10
20
30
40
50
60
|
10
20
30
40
50
60
|
| 交易品种 | 毛利, pips | 损失, pips | 利润, pips | |
|---|---|---|---|---|
| USDJPY | 1.6K | |||
| EURJPY | 591 | |||
| GBPJPY | -2.4K | |||
| GBPUSD | -540 | |||
| AUDJPY | 231 | |||
|
2K
4K
6K
8K
|
2K
4K
6K
8K
|
2K
4K
6K
8K
|
- 入金加载
- 提取
基于有关不同交易商真实账户的执行统计的平均滑移点按点数指定。它取决于 RoboForex-Pro 提供商以及订阅者之间不同的报价,以及订单执行的延迟。值越低意味着复制的质量越高。
|
FPMarketsLLC-Live
|
0.00 × 1 | |
|
VantageInternational-Live 6
|
0.00 × 16 | |
|
PhillipNova-Server
|
0.00 × 2 | |
|
SCFMLimited-Live2
|
0.00 × 23 | |
|
PlexyTrade-Server01
|
0.00 × 4 | |
|
VantageInternational-Live 3
|
0.00 × 7 | |
|
xChief-MT5
|
0.00 × 1 | |
|
CudraniaCapital-Real
|
0.00 × 1 | |
|
STMarket-Live
|
0.00 × 4 | |
|
DerivSVG-Server
|
0.00 × 14 | |
|
ICTrading-MT5-4
|
0.00 × 514 | |
|
BeirmanCapital-Server
|
0.00 × 1 | |
|
Deriv-Server-02
|
0.00 × 1 | |
|
ICMarkets-MT5-2
|
0.00 × 84 | |
|
XMAU-MT5
|
0.00 × 1 | |
|
HedgeHood-MT5
|
0.00 × 75 | |
|
OnePrime-Live01
|
0.00 × 1 | |
|
SolidECN-Server
|
0.00 × 8 | |
|
EverestCM-Live
|
0.00 × 16 | |
|
ImperialMarkets-Live
|
0.00 × 2 | |
|
FinexBisnisSolusi-Real
|
0.00 × 135 | |
|
ACYSecurities-Live
|
0.00 × 35 | |
|
Tradestone-Real
|
0.00 × 10 | |
|
Exness-MT5Real17
|
0.00 × 4 | |
|
Exness-MT5Real29
|
0.00 × 7 | |
Donchian-Breakout
Overview
This algorithmic trading strategy is a price action-based breakout system using Donchian Channels to define key high/low breakout levels. The system is built and validated using over 30 years of historical Forex data, specifically designed for the H1 timeframe. Robustness is ensured through rigorous multi-layer validation, including Out-of-Sample (OOS) testing, Monte Carlo simulations, Holdout testing, and Walk Forward Analysis.
Automated Strategy Generation (Randomized, Not Genetic)
Unlike strategies developed through genetic or evolutionary algorithms, this system is constructed using a fully randomized logic generation process:
-
Entry/exit rules, signal filters, and conditions are randomly combined within predefined structural boundaries
-
This approach avoids overfitting commonly seen in genetic algorithms and allows for greater logic diversity
-
Only strategies that pass extensive robustness filters and perform well across multiple symbols are retained
Each week, over 100 million randomized strategies are generated and tested on a selected Forex pair.
During the build phase, each candidate strategy is also validated on 9 additional Forex symbols to ensure cross-market reliability and eliminate symbol-specific curve fitting.
Breakout Entry Logic via Donchian Channels
The strategy identifies breakout levels based on Donchian Channels, which track the highest high and lowest low over a specified number of bars:
-
A Buy Stop is placed slightly above the Donchian Upper Band
-
A Sell Stop is placed slightly below the Donchian Lower Band
-
An ATR-based pip buffer is added to both directions to reduce false triggers due to short-term volatility
The Donchian period (e.g. 20–80 bars) is dynamically selected during strategy generation and lightly optimized within constrained ranges.
Entry Conditions: EMA Filter
To improve breakout quality and filter out noise, the strategy requires an EMA-Filter to be valid before any Stop Order is placed.
Risk Management & Trade Protection
Every trade is executed with strict risk controls, including:
-
Stop Loss (SL) and Take Profit (TP) — often dynamically based on ATR or Donchian range
-
Trailing Stop or Break-Even logic (with buffer) to protect profits
-
Automatic Friday Close to avoid weekend gaps and exposure
Optional risk features:
-
Time-of-day filters
-
Trade frequency caps
-
Maximum open trades
-
Session-based filters
Robustness & Validation Process
To ensure high confidence and long-term stability, each strategy passes through extensive testing procedures:
-
Out-of-Sample (OOS) validation on 30–40% of data
-
Monte Carlo simulations (random order delay, slippage variation, data shuffling)
-
Walk Forward Optimization and Validation
-
Cross-Market Testing during the build phase (on 9 additional Forex symbols)
-
Timeframe robustness checks (for generalization across market types)
Execution Parameters
-
Timeframe: H1 (Hourly)
-
Markets: Up to 10 Forex Major and Minor Pairs (still in Building Phase)
-
Entry Type: Stop Orders at Donchian Channel breakouts with ATR-based pip buffer
-
Execution: Fully automated mit MT4 VPS
-
Strategy Logic: Randomly generated logic blocks under strict robustness and structure constraints