Stock Prediction Using RNN
Overview
High Frequency Trading Price Prediction using LSTM Recursive Neural Networks
In this project we try to use recurrent neural network with long short term memory to predict prices in high frequency stock exchange. This program implements such a solution on data from NYSE OpenBook history which allows to recreate the limit order book for any given time. Everything is described in our paper: project.pdf
Program is written in Python 2.7 with usage of library Keras – installation instruction To install it one may need Theano installed as well as numpy, scipy, pyyaml, HDF5, h5py, cuDNN (not all are actually needed). It is useful to install also OpenBlas.
Author: Karol Dzitkowski
Dependencies
- Python 3.5
- Numpy
- keras
- tensorflow
MIT License (MIT)
How to run
- cd StockPredictionRNN
- cd src/nyse-rnn
- mkdir symbols
- python nyse.py
- python main.py
