INVESTING — DATA SCIENCE AND ANALYTICS
Supercharge Technical Analysis with Machine Learning
A comprehensive guide to adding Machine Learning layer to Technical Analysis for stock trading — Using PyCaret and Ta-Lib
Where we left off.
In my last article on the subject titled Automating Technical Analysis for Stock Trading With Python, we discussed how to:
- Define the universe using Top N stocks based on Market Cap from our choice of index
- Use the TA-Lib library to find if any of the 61 Technical Chart Patterns show up in the universe of stocks we selected
- Assimilate the Bullish and Bearish pattern signals into a unified score to find Top-Buy and Top-Sell ideas for the day
We also discussed that the approach had two major shortcomings:
- It weighs all patterns equally and does a linear aggregation of signals. In real life, different chart patterns will have different levels of importance
- It omits the interaction effect between patterns and patterns + indicators. Specific chart patterns only become relevant when volumes support them or where the prices are from the moving averages etc.
A technical analyst would generally apply heuristics and collection of personal experience to weigh the charts and interactions thereof to generate buy/sell signals and their respective probabilities. However, this does not have to depend on individual memory and heuristics. We can use Machine Learning to do that work for us — objectively and efficiently. This is what we will try to achieve in this article.
To do this we would use the ultra-efficient PyCaret. PyCaret is a low-code machine-learning tool that has reduced the time and effort of experimentation for citizen as well as professional Data Scientists by several orders of magnitude.
![Riazuddin Mohammad](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="40" height="40"><rect fill-opacity="0"/></svg>){kind=small}
