InfoQ attended the Jax Finance conference held in London on April 4-5, 2017. Jax Finance is a vertical event, co-located with Jax DevOps. Speakers included Stephen Colebourne of Joda fame, Pierce Crosby from StockTwits, and Eric Horesnyi from StreamData.io.
Stephen Colebourne presented on Strata, the Open Source Java Library for Market Risk. Strata is one of only two open source market risk offerings (the other is QuantLab, written in C++).
Starta is written in Java version 8 and is licensed under the Apache v2 license. Installation is straightforward - it uses just JAR files with no servers or database required, and is released on Mavern central. The API looks impressively well designed and easy to use, and there are lots of code examples available.
The decision to use Java 8 was taken since it has a number of features that are very beneficial for Market Risk Analysis - Colebourne specifically mentioned Date and Time, Streams and Lambdas, and Methods on interfaces. It results in an impressively clean API.
Trades are kept as immutable beans, and pricing/risk logic is stateless, and separate from trades. The product has three levels of pricing/risk API:
Pricer - one trade, one set of market data
Measure - one trade, one set of market data
Calc - many trades, one or many sets of market data
It has an impressive range of asset class coverage, with lots of built in conventions.
For instance, Colebourne walked through an example using a Forward Rate Agreement (FRA), a financial instrument that represents the one off exchange of a fixed rate of interest for a floating rate at a future date. A FRA is represented in Strata using the Fra class, and can be created using a builder, as follows:
The trade details are kept in a FraTrade class:
Since most OTC trades follow market conventions, Strata includes definitions of some of these conventions including FRAs,18 different ibor-like indices such as GBP LIBOR, 12 overnight indices such as GBP SONIA (Sterling Over Night Index Average), and currencies and basic holiday data, although Colebourne stressed that the holiday data was intended for evaluation purposes rather than prodcution.
For example, here is a sample code Colebourne showed to create a FRA using a convention:
And here is an example of creating a SWAP from a convention:
Commercial support for Strata is available from OpenGamma, and it is also the basis of their commercial offerings.
Elswhere at the event, Pierce Crosby spoke about StockTwits, a niche social network for investors and traders, primarily focused on equities. The core function is to provide a chat for people who are investing in the markets.
The platform is written in Ruby with a front-end in responsive HTML with an iOS and Android app.
It is much bigger than you might imagine:
1,000,000+ registered users
60,000,000+ trading idea
9,7000+ assets covered: equites, futures, currencies, commodities, indexes and ETFs
Eric Horesnyi, who was also speaking at the event, mentioned StockTwits in the podcast we recorded with him.
