Machine Learning for Stock Trading
- English
- Artificial Intelligence
Machine Learning
- Project length: 8h 16m
Have you ever wondered if AI could earn tremendous amount of money on the stock market and outsmart other human investors? An AI-powered trading robot can automatically make trade decisions on your behalf and allow you to profit from the market without your direct involvement. In this project tutorial, you’ll learn how to use machine learning to develop a stock trading robot. You’ll gain all the essential skills to create a full-fledged stock trading algorithm that investors and traders can utilize in their trading.
Overview
Introduction
In this project, we’ll start by developing an application that reads price data from an API and saves it on a MongoDB database. Another part of this application will be using the saved data for training a machine learning model, optimizing, and then saving it. The last part will be using the saved model and applying it to real-time data provided by the brokerage company while trying to predict the stock price movements. It’s that exciting!
What are the requirements?
You’ll grasp the concepts described in this project quickly if you have some basic skills in the following:
- Python 3.5+
- Database knowledge (understanding of MongoDB or NoSQL)
- Mathematical skills
- Machine learning tools and technologies such as Anaconda, Jupyter, Keras, Scikit-learn, and TensorBoard
- And the most important is: desire to learn
Who is the target audience?
- Do you want to build an automated trading system?
- Do you want to learn more about the financial markets?
- Do you want to learn more about machine learning and neural networks?
- Do you want to know how AI can be applied to investing?
What will you learn after finishing this project?
- How to create trading robots for executing orders on the stock market automatically
- How to apply algorithmic trading using probabilistic machine learning techniques
- How to create an application that accesses live market data, evaluates it, and decides whether to place trades
- How to optimize a trading robot for profitable trading in the financial markets
Project Outline
Session 1: Init/Clone Github Repo; Create a Broker Account
- Setting up the Python environment for Windows
- Setting up the Python environment for Linux
Session 2: Read Data from API, save Data to MongoDB
- REST API
- MongoDB
Session 3: Prepare data for the model, define and extract Y
- Pandas & Scikit-learn
- Prepare raw data to be "digestible" by our machine learning model
Session 4: Define the type of ML model and prepare model viz
- Keras
- TensorBoard
Session 5: Train and Save Keras Model
- Optimize params
- Save model
Session 6: Predict the Outcome of New Incoming Stock Data
- Predict-worker
- OandaV20 API Price Pull (streaming_data collection in MongoDB) Use saved model