Abstract
Changes in intraday trading volume are integral to any algorithmic trading strategy. Accordingly, forecasting the change in trading volume is paramount to better understanding the financial markets. This paper introduces a new method to forecast the log change in trading volume, leveraging the power of Long Short Term Memory (LSTM) networks in conjunction with Support Vector Regression (SVR) and Autoregressive (AR) models. We show that LSTM contributes to a more accurate forecast, particularly when constructed as part of a hybrid model with AR. The algorithm is extended to include data about the time of day, helping the model associate the log change in trading volume with the current hour, which yields the best performance of all trials.
Highlights
In recent years, deep learning became the subject of a growing body of research in many disciplines, including applications in finance (Dixon et al, 2017)
Our results show that Long Short Term Memory (LSTM) contributes to a superior prediction of the change in volume
As can be seen from the table, LSTM-AR-HR gave the best performance of all models, with Mean Absolute Error (MAE) of 0.7669 and correct direction of 0.7054
Summary
Deep learning became the subject of a growing body of research in many disciplines, including applications in finance (Dixon et al, 2017). LSTM networks in particular demonstrated success in natural language processing as well as in predicting the element in a sequence or even the entire sequence This ability can be applied to prediction of financial trends, including change in trading volume of stocks—a subject with high significance as it can be applied to assist in solving a wide variety of financial problems. A trader may decide to limit intraday exposure, e.g., exposure throughout the trading day, in accordance with changes in trading volume. This area of research may have some applications in regulatory settings. A model that can predict the change in trading volume may be useful in recognizing irregular activity, such as a sharp increase in volume when a decline would be expected
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call