multivariate_hawkes_lob

This repository contains the code used to collect Limit Order Book data from Bitfinex, convert them into a Pandas dataframe, and use them to train and simulate Multivariate Hawkes Processes.

Requirements

You need to have Conda or Miniconda installed on your system.
If you don’t have it yet, follow the official installation guide:
https://docs.conda.io/projects/conda/en/latest/user-guide/install/

Setup

You will create two conda environments — one for data collection and one for training/simulation.

1) Environment for collecting LOB data

Create the environment from environment_bitfinex_api.yml:

conda env create -f environment_bitfinex_api.yml

2) Environment for training & simulating Multivariate Hawkes Processes

Create the environment from environment_hawkes.yml :

conda env create -f environment_hawkes.yml

---

Usage

A) Collect Limit Order Book data

Activate the data environment and run the collection script. Then convert the raw data to a Pandas dataframe.

conda activate bitfinex_api
python collect_lob_data.py
python convert_lob_data.py

B) Train Multivariate Hawkes Processes

Activate the Hawkes environment and run the training script (greedy algorithm):

conda activate hawkes
python train_multivariate_hawkes_with_greedy_b_all_training_periods_bi.py

C) Simulate events

Still in the Hawkes environment, simulate events:

python predict_mid_price_events.py