Forex Automated Trading Tool [Dow Theory Swing Trade Type EA Selene] Product Overview

Contents

目次を全て見る

Product Overview
The method of use is simple!
Performance (backtest, simple interest)
Performance (forward test, simple interest)
Performance (back-tested and compounded)
Performance (forward test, compound interest)
Real forward results (compounded)
operating environment

Product Overview

Sequential swing trade type EA based on Dow Theory

Dow Theory MT5 Indicator Increases Forex Winning Ratio!

Dow Theory Indicator developed! We have developed an MT5 indicator that applies the Dow Theory approach. (I spent several months developing one indicator...) We are pleased to intr

The Dow Theory Indicator will make entries in the forward direction at points judged to be favorable for entry based mainly on trend direction judgment and waveform recognition for each of the long, medium, and short term trend indicators. This EA works well with stocks and time frames that tend to continue trends.

The image above shows the Dow Theory indicator superimposed on the EA's backtest results.　The EA itself does not have the function to draw various lines and objects on the chart.

Machine learning is utilized to achieve a 75~80% win rate!

Various indicators are used to evaluate the strength/weakness of trends and volatility, and these results are scored via machine learning to enable complex decisions that cannot be expressed using simple rules-based logic. This dynamically determined the market environment that was considered favorable for entry and raised the win rate.
Although there are variations from issue to issue, the back and forward test results show that the 9 issues covered by the release have a win rate of approximately 75~80%.
The back-test and forward-test results show that the 9 stocks in the release have a win rate of approximately 75~80%.
The back and forward test results are shown in the following table.
The machine learning behavior required a large number of parameters, and even the optimization of a single issue became extremely time-consuming.

Functional design with emphasis on stability!

Moderate number of entries

Winning market conditions do not come often. Instead of forcing an entry, we carefully select favorable market conditions for entry.
The back-test results show that even with 9 stocks operating simultaneously over 180 months, the number of trades was 7,406. The monthly average is about 4.5 times per stock, which is considerably less than that of a typical automated trading tool.
While the number of entries is low, the highly versatile logic allows us to operate in multiple stocks, and we have tried to increase operational efficiency through portfolioization.
No more than 1~2 simultaneous positions per stock will be held at the same time in most cases.　No unplanned pick-and-pin to cover losses.

Automatic adjustment of stop-loss limits

After a position is held, the stop loss is automatically adjusted in conjunction with a new low during a rising Dow trend (i.e., a new push low) or a new high during a falling trend (i.e., a new return high).

Trading style less sensitive to spreads and delivered rates

Since it is not a scalping type of trading where the trader repeatedly targets narrow price ranges, the results do not fluctuate greatly due to slight differences in spreads or tick delivery rates.

Responding to the risk of sudden market changes

It reads the economic indicator calendar and dynamically suppresses entries before the release of important indicators and adjusts stop-losses on open positions.

Can be operated from a small amount!

Funds management mode

It can be operated in simple or compound interest mode.
For compounding mode, a fixed margin method is used. This method dynamically manages the number of lots according to the surplus margin, leverage, and risk % of the starting parameters. Since the lot size is dynamically controlled taking into account the amount of funds and the state of unrealized losses/gains, it is not only easy to increase profits when winning in a row, but it is also expected to reduce the amount of losses when losing in a row.

portfolioing

The EA is optimized based on the rate characteristics of each stock/time frame/trading direction.　Diversification of stocks and time frames improves overall stability and operational efficiency.
Although the frequency of trading per EA is much more modest than that of general automated trading tools, the versatile logic that can be adapted to a wide range of stocks ensures many trading opportunities by operating with multiple stocks.

release notes

Date	Contents
2024/04/30	v1.0 Add new issue EURNZD version (20 minutes)
2024/04/16	v1.0 released USDJPY version (10 minutes) GBPJPY version (20 minutes) GBPUSD version (10-minute) USA 100 version (20 minutes) AUDJPY version (20 minutes) XAUUSD version (5 minutes) AUDUSD version (20 minutes) USDCHF version (20 minutes)

The method of use is simple!

At startup.
You can start trading immediately by adjusting the money management mode (simple or compound interest) and the number of lots.

The initial parameters are pre-optimized and embedded for each issue, so that you can start using the system without having to perform difficult tasks such as analyzing and optimizing the parameters by yourself.

The total number of parameters is about 150 as a result of mimicking the computational process of machine learning in the development of EA, and it is not realistic for a human to set each parameter individually in his/her head.

Parameters that do not significantly degrade generalization performance are adopted after optimization for each stock and sufficient forward test evaluation to avoid over-learning.

Performance (backtest, simple interest)

test conditions

Test period 15 years (2008/01/01~2022/12/31)
Initial margin 100,000 yen (Initial margin is set at 200,000 yen only for the combined results of all issues)

summary

Outstanding Balance	Monthly Drawdown Trends

description	total profits or losses	Margin Maximum DD	PF (Profit Factor)	RF (Recovery Factor)	SR (Sharpe Ratio)
USDJPY	347,993 yen	31,524 yen	1.37	11.04	1.66
GBPJPY	556,236 yen	56,061 yen	1.48	9.92	1.58
GBPUSD	538,765 yen	65,199 yen	1.33	8.26	1.70
USA100	443,605 yen	44,090 yen	1.84	10.06	3.15
AUDJPY	258,222 yen	21,368 yen	1.70	12.08	2.74
EURNZD	298,766 yen	29,340 yen	1.71	10.18	1.78
XAUUSD	234,842 yen	28,375 yen	1.65	8.28	3.20
AUDUSD	182,072 yen	18,743 yen	1.78	9.71	3.82
USDCHF	145,373 yen	18,511 yen	1.63	7.85	2.67

Test Results by Brand Name

description	Outstanding Balance	Test Results
USDJPY
GBPJPY
GBPUSD
USA100
AUDJPY
EURNZD
XAUUSD
AUDUSD
USDCHF