Stock prices look very much like random walks: the signal-to-noise ratio is close to zero. Therefore, trying to model the prices directly to make investment decisions is extremely challenging. A different approach is to predict a company’s fundamental financial data (revenue, assets, debts, and so on). If you can know one year in advance what the financial health of a given company will be, then it obviously becomes much easier to make good investment decisions.
As was shown in “Improving Factor-Based Quantitative Investing by Forecasting Company Fundamentals,” a recent paper by John Alberg and Zachary C. Lipton presented at NIPS 2017, good predictions can be made using deep learning—more specifically using LSTM recurrent networks. Aurélien Géron explains how to forecast stock prices using the solution proposed in that paper, which is an excellent example of using LSTMs for time series forecasting. In the process, he details how LSTMs work, their advantages and limitations, and some alternatives. You’ll also learn how to perform cross-validation when dealing with time series, in particular to search for good hyperparameters.
Aurélien Géron is a machine learning consultant at Kiwisoft and author of the best-selling O’Reilly book Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow. Previously, he led YouTube’s video classification team, was a founder and CTO of Wifirst, and was a consultant in a variety of domains: finance (JPMorgan and Société Générale), defense (Canada’s DOD), and healthcare (blood transfusion). He also published a few technical books (on C++, WiFi, and internet architectures), and he’s a lecturer at the Dauphine University in Paris. He lives in Singapore with his wife and three children.
Comments on this page are now closed.
©2018, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org