What is difference between machine learning strategies and without machine learning strategies

AI & Machine learning
1

Could you please elaborate?  what is difference between machine learning strategies and  without machine learning strategies ?  what is benefit of useing machine learning strategies

2

Hello Shirish,



You may agree that to create trading strategies, you need to have a set of well-defined rules.

These rules help primarily to ensure that you are trading systematically and are not driven by emotions. 



Now the question that arises is, how can you define these rules?



The answer is really simple!

The trading rules can be created using conditions as simple as the following:

If the current price is greater than the previous day's high, then buy the asset.



On the other hand, you can also introduce indicators (For example, Moving averages, RSI, etc) to dynamically generate signals, in order to create a much more refined strategy.



But what if you want your strategy to be a bit more complex than this?



Imagine having a strategy that can analyse heaps of sentiment data gathered from news or tweets and forecast the possible future price movements based upon it.



Similarly, how about passing some feature data to a strategy?

For example, data of several technical indicators, fundamental data, macroeconomic data, etc and waiting for the strategy to reveal some hidden patterns in it for you?



Also how about leveraging the power of Artificial Intelligence to conveniently solve complex problems such as capital allocation, risk management, and much more for your existing strategies/portfolios?



You can achieve all of this and much more with the help of Machine Learning.



To learn more about the applications of machine learning when it comes to trading, I encourage you to read some interesting blogs at QuantInsti. You can start off right here.



I hope this was helpful to you.

All the best, Shirish.

3

Thanks  My confusion is clear, 



in machine learning can we ideditfiy high probability enter and exit signals? 



if yes  please share examples

4

Hi Shirish,



By high probability entry and exit signals, I'm assuming that you are referring to the use of machine learning algorithms for generating trading signals that turn out to be accurate for the most part.



If you think about it from a different perspective, it's as good as asking:

Can we identify high-probability trading signals by using certain technical indicators, entry-exit rules, etc?



Of course, you can get a few good trades here and there.

But, the key to having good accuracy depends on multiple factors like:

  • The technical indicators or the trading rules you choose to work with.
  • The way you interpret these indicators.
  • The very conditions you set for the entries and exits.
  • Your overall strategy logic, execution plan, etc.



    And finally, how well you put together and implement all of the above.



    Similarly, while a machine learning algorithm can help in many ways, it eventually boils down to:
  • what data you feed it with, the complexity of the output you are looking for, and many other factors.



    A quick example, 

    Think about Youtube's algorithm that constantly tries to recommend content that you may find interesting. For the most part, you may notice that it does a pretty decent job. But occasionally, you may also find some suggestions that are not to your liking.



    Now does this mean that the algorithm is not working correctly?



    Not at all! It simply means that the algorithm still has scope for improvement and can be trained more in order to give better recommendations (or in other words, to improve its accuracy).



    So to finally answer your question, yes you can definitely use machine learning to identify high-probability entry and exit signals. But it all depends on your individual capability to train the machine learning model in the best way possible.



    I hope this gives you some clarity on your doubt.
5

Thanks Kevin ,  it very usefully information and  clear my all doubt…