Hi, I'm Jack. One of the course creators and also an experienced trader now working with the New York Institute of Finance. Welcome to this course series all about machine learning and trading using Google Cloud. In the first course, we covered the functionality of Google Cloud and Jupyter Notebooks, along with introducing basic quantitative trading strategies, supervised learning, neural networks, and deep learning. In this course, we'll explore more advanced concepts and trading such as momentum and pair trading strategies. You'll also learn how to write low level programs in TensorFlow and use Keras to train neural networks. In course number 3, you'll deepen your understanding of advanced machine learning approaches such as reinforcement learning, which has unique advantages when used as the foundation for a quant trading system. Machine learning for trading is a broad topic, so let's first be clear about who can benefit from this course content. This course was created in partnership between the New York Institute of Finance and Google Cloud. It's a combination of machine learning theory specific to finance and trading and the practical application of that theory using Google Cloud as a platform. Before you take this course, it's recommended that you have already taken a foundational machine learning course, so you'll be familiar with the concepts discussed like feature engineering, training, serving splits, and more. I'll provide links in the course resources of courses and reading materials for you to get caught up to speed. The audience that will get the most out of this series of courses are data analysts, data scientists, and data engineers who are looking to apply their ML knowledge to the finance domain, specifically trading. We'll cover and quiz you on core trading concepts like fundamental versus technical trading, as well as quant theory. Portfolio managers and traders can also use his course to learn how machine learning techniques can be applied to investment in trading decisions. Lastly, while you won't leave this course building the next high-frequency trading model for Nasdaq 100 futures, you'll leave with a better understanding of how you can apply financial trading concepts in theory using the latest in Cloud technology. Now let's cover the specific module content that you'll learn in this course. There five modules. Module 1, quant trading strategies. Two, introduction to TensorFlow. Three, training neural networks with TensorFlow and Keras. Four, building a momentum base trading strategy and lastly, in five, you'll build a pair trading strategy prediction model. A lot of learners asks what they're expected to code is part of this course. While there are eight interactive IPython Notebook based labs for this course, you'll develop your skills with TensorFlow and Keras in the four lab shown here. One, writing low-level TensorFlow programs. Two, manipulating data with TensorFlow dataset API. Three, introduction to Keras sequential API, and for, intro to Keras functional API. You'll also develop basic momentum and pair trading strategies in this three labs. Momentum trading, pairs trading, and Kalman filters. We're also making available an optional lab where you can try to improve your momentum strategies using Hurst coefficients. You'll then have the option to take additional momentum labs using ML in the QuantQuest platform from Aquant. Through our partnership with Google Cloud, each of your labs would be using a real Google Cloud account at no additional cost and compute resources to train and run your models. If you've used Qwiklabs before, you've seen the lab environment shown here. You'll be given a certain amount of time to use the project resources before they expire and you'll follow along with the lab instructions in the Qwiklab while doing the actual work in your provided account. Since most of these labs are machine learning-based, you can expect to see IPython Notebooks like this one that you'll be reading through and executing. If you're unfamiliar with IPython Notebooks, check the course resources for a quick primer or look at the lab solutions where we go through all the lab steps. One key point to note here is that all of these notebooks are available publicly in our course repository linked here and in the course resources. That means even after this course ends, you can still bookmark and refer back to the code for this course and future courses. Just as is critical, go over what exactly this course covers is equally important to talk about what we're not going to cover. Machine learning trading and Google Cloud are three very broad topics, and this series of courses navigates their intersection. Naturally, we can't be everything for all audience who will provide links for newer data scientists to get up to speed and advanced challenges for gurus to show off their skills. Specifically, what you won't see here are building and implementing the next highly profitable high-frequency pair trading algorithm. At the core of highly profitable models is likely a series of very private and expensive to collect training datasets that feed into multiple models. This course will teach you the core trading concepts used in professional models, but it's up to you to beat everyone else in the market and build a better model of reality for a given stock. With that same topic, the more advanced machine learning topics such as LSTMs and reinforcement learning models are covered in course number 3. If you already have a deep understanding of ML, you may want to focus on learning the new trading concepts in this course and experiment in the labs and how you can beat the benchmark given your ML knowledge. Lastly, if you're new to machine learning in Python, we won't cover basic intro concepts like what is a feature of model input as there's a wide universe of great generic ML content out there online. Check the course resources for examples, and then come back here when you want to apply that theory to trading. What's next? In the next course; Reinforcement Learning for Trading Strategies, you'll dive into building models with TensorFlow and Keras. One of the specific model types is LSTM or long short-term memory models for better time series prediction. We'll then look at how reinforcement learning techniques can be used to develop a nearly autonomous trading system. Don't forget to review the course resources to access and download the slides for the modules you've just completed, and we'll see you in the next course.
