FX Auto Trading - How to identify when to change strategies dynamically?

[smartautotradingbot]

Hi,

I'm currently working on developing an FX auto-trading system. I came across a video that suggested incorporating multiple strategies can help in reducing risk.

I'm aware of indicators like RSI and MA, which are useful for determining when to adjust position size or the timing of order execution.

But about dynamic strategies. Are there any specific analysis metrics or indicators that can be used for this?


Thank you.

 

[schizo]

You mean like switching between RSI and MA? Sure, it can be done. For example, MA's can be used in a trending market while RSI can be used for rangebound markets.

 

[shine]

You mean like switching between RSI and MA? Sure, it can be done. For example, MA's can be used in a trending market while RSI can be used for rangebound markets.

The problem is that we never know how the market will move today, in a range or along a trend. This can only be seen on the past history of the chart. Therefore, the question arises what to use on a particular trading day, a robot based on MA or RSI?

 

[schizo]

The problem is that we never know how the market will move today, in a range or along a trend. This can only be seen on the past history of the chart. Therefore, the question arises what to use on a particular trading day, a robot based on MA or RSI?

Well, you don't need to wait an entire day to find out if it's a trending day or a rangebound day. Likewise, a smart algo should be able to adapt to the changing environment as the new data come in. Of course, the algo doesn't see the chart in the same way as humans do and will not necessarily draw a clean line between a trend and a range. But it should be able to autonomously change and update the parameters (for example, risk parameters) based on the incoming inputs like volatility.

 

[shine]

Well, you don't need to wait an entire day to find out if it's a trending day or a rangebound day. Likewise, a smart algo should be able to adapt to the changing environment as the new data come in. Of course, the algo doesn't see the chart in the same way as humans do and will not necessarily draw a clean line between a trend and a range. But it should be able to autonomously change and update the parameters (for example, risk parameters) based on the incoming inputs like volatility.

It is clear that a smart algorithm should be able to do this in order to make a profit in any market condition. But how can you create an algorithm that would not drain deposits when the market changes and would work for a long time? That is why many robots work profitably for a very short time.