Analytics for Trading using NOA

In this repository we present a few examples for trading analytics that leverage the nonlinear optimisation platform NOA and its experimental kotlin-jvm frontend within the KMath library.

Installation

To use kmath-noa, you will need first to build & publish the module locally.

For data storage, you have to install MongoDB. You can edit the configurations in the conf.yaml file.

High Frequency Trading in Cryptocurrency Markets

In this study, we explore the impact of High Frequency Trading on execution costs and liquidity for several Cryptocurrency exchanges.

You will find the data collection utilities within crypto-hft-data. To run the app you need to simply execute:

./gradlew -Dorg.gradle.java.home=/path/to/local/jdk -q :crypto-hft-data:run --args=/path/to/conf.yaml

You can configure it in conf.yaml to get the LOB data feed from the following exchanges and liquidity aggregators:

for any of the traded instruments available on the platforms.

Once you've collected a bit of data you can run the analysis with the models we provide in crypto-hft-analytics.

The module crypto-hft-visual provides the visualisation apps.

Acknowledgements

The following research has been carried out using our platform:

R. Grinis, A. Kislitsyn, I. Drozdov and K. Shulga Are Cryptocurrency Markets Running Behind the Fed? A Significant Shift in Crypto Markets Microstructure (October 20, 2021). Available at SSRN

We kindly acknowledge support from Finery Markets, GrinisRIT ltd. and JetBrains Research.