Machine Learning for Traders

In the following video, we will see that the foundation of machine learning is statistics. We will also explore the differences between formulas, classic statistics, and machine learning. Understanding these differences is essential for building a solid foundation in machine learning.

Key Elements of Machine Learning: Essential Suitcases for the Journey

The machine learning pipeline involves a series of essential steps to put a model into operation, starting from the initial planning stage until the model is finally evaluated.

In this video, we will learn about Supervised and Unsupervised Learning, focusing on predicting a class or a number.

In this video, you will learn about constructing your own technical indicator.

We aim to merge multiple correlated technical indicators into a single one, retaining as much signal in the data as possible.

In this video, we will explore two big pictures. First, we will look at the different tasks that can be performed with machine learning. And second, we will see various machine learning algorithms categorized under supervised and unsupervised learning, deep learning, and reinforcement learning.

In this video, you will learn about the crucial role of data in machine learning pipelines. We'll explore the distinctions between tabular and unstructured data, and how they are used in both classic and deep learning models. Classic learning has limitations in the types of data it can use, while deep learning has no such limitations.

In this video, you will learn about the crucial concepts of Explanatory Data and Predictive Data in the context of machine learning for trading.

In this video, through various cases and examples special for traders, you will learn about explanatory data. Culminating in a comprehensive understanding with a Big Picture Guide.

In the previous lecture, we saw the importance of  predictive data, ending with a  Big Picture Guide. In this lecture, we will walk through this map, exploring each of its areas in detail. We will delve deeper into Explanatory Data.

To make the SMA more useful and robust for all traders, it is necessary to normalize it, to ensure that the learned rules are applicable across different price scales. In this lecture we will delve deeper into this topic.

K-Means clustering is a widely used algorithm for partitioning datasets into distinct groups based on feature similarity. To ensure optimal performance, certain preprocessing steps are essential. This lecture covers all the necessary considerations.

In this video, you will learn about what data can be predicted, which is key to unlocking the power of machine learning in trading. Often, we underestimate the critical importance of realistically predictable data in the financial markets domain.

Overcoming Imbalanced Predictable Data for trading.

In this video, you will learn what to do when the date we want to predict is imbalanced.

Imbalanced data occurs when some classes are significantly underrepresented compared to others in a dataset. For example, we might have many days where the daily return was up and only a few where the daily return was down.

Algorithms struggle to generalize information effectively in such scenarios, often favoring the majority class. Machine Learning is powerful but lazy, and it will always look for a shortcut if possible; favoring the majority class is its shortcut!

This lecture outlines the specific requirements for various machine learning tasks and algorithms when it comes to our predictable data.

In this video, we will Discuss a practical case, Clustering S&P 500 components.

We will see the goal,  understanding what we obtain by grouping companies.

We will explore what data  we can use and learn how to apply an algorithm to the data.

Then, we will train a model and evaluate it and finally we will modify the model to obtain better results.

In this video we will show you how to access Google Colab and run our python scripts in a safe mode.

1. Understand Cloud-based Python Execution: Learn how to run Python code in the cloud using Google Colab, eliminating the need for any local installations. This makes it accessible and easy to experiment without fear of breaking anything.

2. Simplify Coding Experience: Realize that working with Python in Google Colab is straightforward and user-friendly, making it suitable even for beginners.

3. Advanced Python Utilization: For those already familiar with Python, explore the deeper functionalities and capabilities available in Google Colab, allowing for advanced coding and data analysis.

By the end of this session, you will be comfortable running Python code in Google Colab, understand the benefits of cloud-based coding, and be able to leverage this platform for both basic and advanced programming tasks.