We develop a large-scale deep learning model topredict price movements from limit order book (LOB) dataof cash equities. The architecture utilises convolutional filtersto capture the spatial structure of the limit order books aswell as LSTM modules to capture longer time dependencies.The proposed network outperforms all existing state-of-the-artalgorithms on the benchmark LOB dataset [1]. In a morerealistic setting, we test our model by using one year marketquotes from the London Stock Exchange and the model deliversa remarkably stable out-of-sample prediction accuracy for avariety of instruments. Importantly, our model translates well toinstruments which were not part of the training set, indicatingthe model’s ability to extract universal features. In order tobetter understand these features and to go beyond a blackbox model, we perform a sensitivity analysis to understand therationale behind the model predictions and reveal the componentsof LOBs that are most relevant. The ability to extract robustfeatures which translate well to other instruments is an importantproperty of our model which has many other applications.
To View the Base Paper Abstract Contents
Now it is Your Time to Shine.
Great careers Start Here.
We Guide you to Every Step
Success! You're Awesome
Thank you for filling out your information!
We’ve sent you an email with your Final Year Project PPT file download link at the email address you provided. Please enjoy, and let us know if there’s anything else we can help you with.
To know more details Call 900 31 31 555
The WISEN Team