[See Description] Creating our Machine Learning Classifiers - Python for Finance 16

TL;DR

This tutorial discusses the process of creating labels for supervised machine learning models in finance, using Quanto Pain and Python.

Transcript

what is going on everyone welcome to part 15 of our finance with Python tutorial series using quanto pain and zipline in this tutorial we're just gonna be building on the last tutorial so if you are lost up to this point please go back to the previous tutorial and ask questions because it's only gonna compound on top of this so we've got our featur... Read More

Key Insights

• 🎰 Supervised machine learning models in finance require both features and labels for training.
• ❤️‍🩹 Creating labels involves comparing the end price with the start price.
• 🎰 Preprocessing, including scaling, can improve the performance of machine learning models.
• ❓ Random Forest Classifier is a popular choice for combining weak classifiers and obtaining reliable results.
• 🌲 The number of estimators or decision trees in the Random Forest Classifier can be adjusted to balance processing time and accuracy.
• ⚾ Printing predictions helps evaluate the model's performance and make trading decisions based on the predictions.
• 🎰 Quanto Pain and Python provide a powerful framework for implementing machine learning models in finance.

Questions & Answers

Q: What is the purpose of creating labels in supervised machine learning?

Labels are used to determine the outcome based on a set of features. They help in training the machine learning model to make predictions.

Q: How are labels created in this tutorial?

In this tutorial, the label is created by comparing the end price with the start price. If the end price is greater, the label is assigned a value of 1; otherwise, it is assigned a value of -1.

Q: Is preprocessing necessary for machine learning models in finance?

Preprocessing helps in scaling the features and improving the performance of machine learning models. However, in this tutorial, the features are already scaled between -1 and 1 due to their nature as percentage changes.

Q: Why is scaling important in machine learning?

Scaling helps in bringing all features within a similar range, making them comparable and avoiding bias towards features with larger values. It ensures that all features contribute equally to the model's predictions.

Summary & Key Takeaways

• This tutorial builds on the previous tutorial and focuses on creating labels for machine learning models in finance.

• The label is used to determine the outcome based on a set of features, and it helps in making predictions.

• The tutorial also emphasizes the importance of preprocessing and scaling the features for better results.