ideasInvestment / warty_warthog

Using LSTM to forcast stock price (idea#1)

Warty Warthog

The sequence of stock price is a typical time series of data. Due to the cyclical nature of stock price movements, future stock price can be predicted based on past stock prices with time sequence analyse. LSTM(Long short-term memory) is a canonical neural network for making time series predictions.

The origin of data

The data used for the analysis is exported data from the DGideas financial database. For legal reasons, we can't provide the original data for analyze in this repository. But any time-correlated sequence should be able to analyzed in the same way.

Reference

• Deep Learning for Time Series Forecasting

Easter-egg

The name of the repository is a tribute to the Ubuntu 4.04 release(Warty Warthog).

Author

• 王万霖 - "DGideas"<[e-mail]>