Historical Data on-Demand: FX price transparency improvements lead to more efficient ways to model and test algorithmic FX trading

Stephane Leroy, Chief Revenue Officer, at QuantHouse explores why FX trading is on the verge of change and why historical data on-demand is key to modelling and testing algorithmic FX trading strategies going forward.

Historical Data on-Demand: FX price transparency improvements lead to more efficient ways to model and test algorithmic FX trading
Stephane Leroy

Since the first exchanges went fully electronic back in the 1990s, there has been an on-going revolution in trading. At that time, screen-based trading applications were created to meet the needs of users in order to improve reliability, speed and volumes. Those applications rapidly expanded to cover a wide scope of requirements designed to enhance day to day tasks related to market data processing, market research, risk management, order management and clearing and settlement.

More than a decade ago, full end-to-end automatic financial processes like electronic trading, black box trading, algorithmic trading, automatic execution engines and direct market data feeds were concepts only a few leading edge experts were working on. These program-based trading innovators - now known as quants - were already treading the path along which trading technologies would ultimately evolve.

Algo Trading is coming

Equities, futures, and options traders were among the first to embrace this quant trading evolution, while other asset classes such as FX were slower to follow the trend, mainly due to the lack of price transparency. Banks typically generated revenues in FX by taking a “spread” on transactions rather than through commissions, meaning there was less incentive to join the quant revolution.  

Fast forward to 2019 however and we are beginning to see a change in the way the FX markets operate. A new report from Greenwich Associates recently confirmed that approximately 20% of institutional foreign exchange trading volume is now executed via algos, and FX is likely to move in the direction of equity markets, in which “algos” will ultimately account for more than half of all trading volume. Whereas this is a relatively low percentage when compared to equities, it still shows a clear market trend, with the combination of regulation, rapid changes in technology and new trading needs forcing the FX market to evolve at an even faster pace to stay ahead. 

The on-going move towards quant trading is significantly changing how FX markets will be structured going forward. Under the new rules introduced in 2018 by the second Markets in Financial Instruments Directive (or Mifid II), banks now have to explain any mark-ups and costs they impose through those means, with the aim of improving best execution. One way this is done is through the use of a consolidated tape and access to historical data on-demand.

Traders Workshop
Historical market data on-demand has become crucial in order to model and test FX strategies quickly and efficiently

 

FX Consolidated Tape: Open Access to Historical Data 

Today, new FX exchanges are providing a consolidated FX tape which brings much needed transparency in spot FX. Now, individuals and companies alike can view the latest FX rates transacted on interbank markets. This significant change has forced the FX trading community to rapidly adopt quant trading technologies to mirror what has been done in other more mature asset classes such as equities.

As a result, historical market data on-demand has become crucial in order to model and test FX strategies quickly and efficiently. Historically, the research, development and backtesting phase has been a lengthy and onerous process. Market participants must identify data sources, align formats and code to those data sources, allocate storage capacity to download the necessary files and ultimately incorporate into their trading models to assess the viability of their trading strategy. By this time the market has often moved on and the backtesting cycle needs to be repeated. 

Historical data on-demand delivers fast, reliable historical data - on-demand - and allows firms to implement new trading ideas within days rather than weeks or even months. For quant traders, it is critical to be able to rapidly detect patterns in historical data sets in order to develop the right trading models to leverage those market opportunities. This is achieved through the combination of advanced research and backtesting tools coupled with historical data services on-demand. 

In addition to leveraging historical data on-demand, FX trading houses must setup  ‘trading factory architectures’ specifically built for quants where the infrastructure, data and algo trading application layers are seamlessly connected.

The trading landscape has changed significantly in the past few years; it is no longer about how fast your trades are sent, but how quickly your trading strategy can be ready. In order to move away from speed trading to smart trading, you need access to trusted, reliable and consistent data on-demand, to enable traders to spot changes and emerging patterns in the market quickly and evaluate and adjust trading strategies accordingly. Historical data on-demand gives firms a distinctive edge by moving to a much more real-time environment and provides a real breakthrough for the algo trading industry.

Given the increasing opportunities in our FX markets and the complexity of setting up “on-demand data infrastructures,” FX trading houses and financial institutions will look to leverage industry solutions available through advanced fintech providers which will ultimately help them gain an edge when adopting quant trading solutions.