Overview

This strategy is based on the Dynamic Moving Average indicator, combined with Bollinger Bands and RSI for trade signal filtering. It implements a trend following long only strategy. The strategy judges the trend by calculating the change of Heiken Ashi closing price Dynamic Moving Average and compares it with Bollinger Bands to generate trade signals. With the RSI filter, it can effectively identify trend explosive points for trend tracking.

Strategy Logic

The core of this strategy is to calculate the change of Heiken Ashi closing price Dynamic Moving Average. Specifically, it calculates the difference between the current bar's MA and the MA of previous two bars, then multiplies it by a sensitivity coefficient to get the accurate MA change value.

Then this change value is compared with the difference between Bollinger Bands upper band and lower band. If the MA change is greater than the BB difference, it is considered as a "trend explosion". When the explosion is positive, i.e. the MA change is positive, it generates a long signal and green bar. When the explosion is negative, i.e. the MA change is negative, it generates a close signal and red bar.

In addition, this strategy has a RSI filter that only allows long signals when RSI is higher than a threshold, avoiding the risk of trend reversal.

Advantages

• Dynamic MA to effectively track trend changes
• BB as a dynamic indicator combined with MA for better trend explosion identification
• RSI filter avoids false signals from low rebounds
• Long only suitable for persistent bull market
• Flexible adjustable parameters for different products and timeframes

Risks

• Long only cannot profit from downtrend
• Overly dependent on parameter optimization for different products and timeframes
• Unable to capture trend reversal effectively, may lead to large losses
• Improper RSI filter settings may miss trading opportunities
• High sensitivity may generate noisy trades

Risk control methods include: proper parameter tuning for robustness, combining other indicators to judge trend reversal, use only in clear long-term trends, etc.

Optimization Directions

There is some room for further optimization:

• Try different price sources like close, moving averages etc. for better smoothing

• Adjust MA and BB period parameters for optimization across different products

• Try ratio relationship instead of sensitivity coefficient for more intuitive indicator value

• Add other filters like trendlines, volume etc. to improve signal quality

• Develop short strategy based on indicator patterns

• Incorporate stop loss mechanisms for better risk control

Conclusion

Overall this is a relatively stable trend following strategy. It uses dynamic moving average to determine trend direction, BB to identify explosive points, RSI to filter false signals, realizing a long only trend system. But it also has some risks, requiring parameter tuning for different products and timeframes, and inability to profit from downtrends. There is room for further improvements like enhancing signal quality, developing short strategy, adding stop loss etc. to achieve better performance.