Udemy

Data Engineer: Prepare Financial Data for ML and Backtesting

Name: Data Engineer: Prepare Financial Data for ML and Backtesting
Availability: InStock
Rating: 4.6 (117 reviews)

4.6(117)7k studentsBEGINNEREnglish

Korshub's take

What You'll Get

A follow-on course for users who have already pulled financial data and now want to shape it into something usable for machine learning or backtesting. Using the no-code Data Engineer tool, the course covers calculating returns as values, percentages, differences, and volatility, adding time sequences, correlation and co-integration between assets, technical indicators, and filters and conditions for future predictions.

Everything is done through the platform's interface rather than by writing code, though the underlying ideas can also be replicated in Python for those who prefer coding it themselves.

Who It's For

Members of the Crypto Wizards platform who have completed the Data Builder course and want to prepare their pulled data for analysis or ML.

Key Takeaways

Calculating returns, volatility, and correlation without code
Adding technical indicators and time-sequence features
Preparing structured features for machine learning and backtesting

About this course

After learning how to extract financial data using Data Builder, you will naturally be wondering how to make use of all the standard data you have pulled. In this course, you will learn how to structure data in such a way that seemingly mundane data can be transferred into useful information that can give you an edge in the financial markets.

Using Data Engineer, you will be able to:

Calculate returns in terms of values, percentages, differences and absolute moves (volatility)
Add time sequences to your data for predictions in machine learning
Add correlation and co-integration information comparing any columns/features for any assets
Add technical indicators
Add conditions for making predictions about the future
Add filters for removing unnecessary data
Prepare your features for machine learning (although not required for backtesting

You will be able to do all of this without writing a single line of code. However, you will need to be a registered member of Crypto Wizards to take advantage of this material as this course was built to teach users (as requested) how to use the platform. If you are not a registered member, you can still take valuable principles away from this course and perhaps code this yourself using Python or another data science related approach.

See you in class.