Dao Liang Ding / Perfil
- Información
2 años
experiencia
|
2
productos
|
11
versiones demo
|
0
trabajos
|
0
señales
|
0
suscriptores
|
![Dao Liang Ding](https://c.mql5.com/avatar/2024/1/65b21d43-92a7.png)
Dao Liang Ding
I have always believed that an excellent EA (Expert Advisor) should have fewer or even no parameters, as EAs that require frequent parameter adjustments are not suitable for live trading. A low-parameter or parameter-free EA means that it can automatically adapt to market conditions and make appropriate decisions without the need for frequent parameter tuning. This type of EA is usually based on powerful algorithms and strategies, enabling it to perform well in various market environments. In contrast, EAs that require frequent parameter adjustments may need constant parameter tuning based on market changes, requiring traders to spend a significant amount of time and effort monitoring and adjusting. In such cases, live trading often faces more risks, as parameter adjustments may lead to unexpected trading outcomes. Therefore, for live trading, I prefer to develop outstanding EAs with fewer or no parameters, as they can execute trading strategies more stably and reliably, reducing the intervention of human factors and improving the reliability and consistency of trading. In quantitative trading, parameter optimization is based on the assumption that future markets will continue or approximate the current conditions. However, markets are chaotic and illogical. If too many indicators or parameters are used in a model, it can get trapped in an endless vortex of parameter tuning. Excessive parameters can lead to overfitting, where a model performs well on historical data but fails to produce consistent results in actual future trading. The uncertainty and complexity of the market make parameter optimization complex and challenging. Therefore, when conducting parameter optimization, we need to carefully select an appropriate number of indicators and parameters to ensure that the model performs stably and reliably in different market environments. At the same time, we should be aware of the market's disorderliness and illogical nature to avoid overfitting and excessive reliance on parameter tuning. To find the best solution, we can evaluate and optimize the model through statistical analysis, backtesting, and validation. This ensures that our EA can continuously adapt to market changes in live trading and achieve reliable trading results.
![Dao Liang Ding](https://c.mql5.com/avatar/2024/1/65b21d43-92a7.png)
Dao Liang Ding
我始终坚信,一个优秀的EA(专家顾问)应该是少参甚至是无参的,因为经常需要调参的EA并不适合用于实盘交易。一个少参或无参的EA意味着它能够自动适应市场条件并做出相应的决策,而无需频繁地进行参数调整。这种类型的EA通常基于强大的算法和策略,能够在多种市场环境下表现出色。相比之下,经常需要调参的EA可能需要根据市场变化不断调整其参数,这需要交易者花费大量的时间和精力进行监控和调整。这种情况下,实盘交易往往会面临更多的风险,因为参数的调整可能会导致不符合预期的交易结果。因此,对于实盘交易,我更倾向于开发那些少参或无参的优秀EA,它们能够更加稳定和可靠地执行交易策略,减少了人为因素的介入,提高了交易的可靠性和一致性。 在量化交易中,参数优化是基于未来市场将持续或接近当前状况的假设。然而,市场是无序和不合逻辑的。如果在模型中使用过多的指标或参数,将会陷入无休止的参数调整漩涡中。过多的参数可能会导致过度拟合,使模型在历史数据上表现良好,但在未来的实际交易中无法产生一致的结果。市场的不确定性和复杂性使得参数的优化变得复杂而困难。 因此,在进行参数优化时,我们需要谨慎选择适当数量的指标和参数,以确保模型能够在不同市场环境下具有稳定和可靠的表现。同时,我们应该意识到市场的无序性和不合逻辑性,以避免过度拟合和过度依赖参数调整所带来的风险。为了找到最佳方案,我们可以通过使用统计分析、回测和验证来评估和优化模型。这样可以确保我们的EA在实盘交易中能够持续地适应市场变化,并获得可靠的交易结果。
: