Erez Katz, Lucena Research CEO and Co-founder
Using Deep Neural Nets to Forecast Stock Prices
There’s been a lot of buzz surrounding using machine learning to forecast securities. In the below webinar recording, I highlight a unique approach to identifying investment opportunities through the use of deep neural networks.
More specifically, how convolutional neural networks (CNN) provide a compelling approach to classifying time series data in order to project stocks’ impending price action.
What You Can Expect:
A brief overview of deep neural networks (deep nets) and how convolutional neural networks (CNNs) are able to classify images.
How CNNs are used to recognize hand writing with uncanny accuracy (above 99.7%).
A discussion on how the success of computer vision using CNNs for image classification, speech recognition, and object detection can also be applied to tradable securities.
How multiple rounds of trial and error led to a compelling solution. Specifically, what seemed promising on paper and in theory, but didn’t work for us.
Lastly, what specific actions we took to “help” the neural networks learn. How we attempted to overcome data not always conforming to IID (Independently and Identically Distributed data) and the non-stationary nature of the financial markets.
Whether you are an investment professional looking to understand machine learning or a quant with experience in quantitative finance and data science, this discussion has something for you.
Our goal is to give you a glimpse into the considerations that a quantitative investment research team takes into account while attempting to provide enough actionable reference for the portfolio managers to succeed. Enjoy!
Full list of Q&A received during the webinar
Video: How to Apply Deep Reinforcement Learning to Trading
Video: The Journey of Validating Alternative Data Signals
Video: Unique Data Feeds for KPI and Price Forecasting
Blog Post: Capsule Networks or CNNs: Which is Best for Stock Predictions?