Skip to content

Repository files navigation

SpatialLOB-Learning-spatial-properties-of-Limit-Order-Book

SpatialLOB is designed for stock price prediction by exploiting spatial properties of the Limit Order book.

Orderbook micro structure consists of multiple price level on each side of the book, as buy side and sell side. These price levels could have different aggregate order volumes. Order volumes of a price points are changing as the price point receives order updates like add, amend, cancel orders. Following figure shows this spatial changes over multiple orderbook updates. Please refer kaggle notebook for more understanding of this view.

About

SpatialLOB is designed for stock price movement prediction by exploiting spatial and temporal properties of the Limit Order books. SpatialLOB consists of a deep network combining CNN and Stacked GRU.

Resources

Stars

9 stars

Watchers

2 watching

Forks

Releases

Packages

Contributors

Languages