Market Quality Dashboard

The MQ Dashboard allows users to choose a number of parameters for their unit of analysis, such as individual stocks in a marketplace, user-driven stock groupings, event study, customized metric. In future iterations, cross-listed stocks analysis, automated control sample will also be provided.

At the centre of MQD are a wide range of pre-calculated metrics representing fairness and efficiency in markets, collectively known as market quality metrics. The system also has all the significant market design changes adopted in every major market over the last 10 years and enables market quality metrics to be visualised pre and post a particular market design change to provide a quick view of the impact of the change on market quality. If you are considering making a change and its occurred elsewhere MQD lets you study the impact.

Alongside publically available data for every exchange traded market in the world, a Python based language known as Uptick enables users to build their own bespoke metrics as well as incorporate proprietary data into the system. The system also incorporates the R open source statistical package which enables a time-series of relevant metrics to be download for detailed statistical analysis. The whole system operates as SaaS on the Amazon web cloud.

MQD has three main users to date:
1.                   Academics who can get their base level results in a fraction of the time it typically takes,
2.                   Exchanges who can use the tool to enhance the data experience for users and
3.                   Securities Regulators for whom the product solves their stated of objective of ensuring that all market design changes face and pass the dual tests of fairness and efficiency.

You can see a series of case studies investigating different market design changes in the US, Hong Kong and Canadian markets at www.mqdashboard.com under Case Studies (New).  You can also see an array of other market design changes being investigated at our YouTube channel.

Financial Market Data ETL Workflow Engine

On a daily basis, the CMCRC extracts, transforms and loads (ETL) financial market data for over 50 global markets, which serves as the vital input to the research program and product development such as the Market Quality Dashboard. In order to maintain processing efficiency and data quality at high level, the CMCRC developed a unique ETL Workflow Engine that has automated data management and outlier detection to the maximum extent. The ETL Workflow Engine seamlessly supports cluster and cloud computing.

Natural Language Processing

Traditional finance research has primarily been based on structured numerical data mainly due to the difficulty with mining unstructured text. The CMCRC developed CompNews aiming to unlock knowledge from this via entities - the companies, people, places and locations that the text is about. The system provides a “computable” layer of structured data over unstructured text, giving user a platform on which to build advanced analytical applications for information leakage detection and research on social media’s impact on financial markets.

Alluvial Backtesting Platform

Alluvial is a new software product from the Capital Markets Cooperative Research Centre (CMCRC) offering algorithmic back-testing and trading solutions to advance the research, development, implementation, and evaluation of automated trading strategies. Financial markets are becoming increasingly technology driven. High frequency and algorithmic trading not only overwhelmingly reforms the finance industry, but also poses evolving challenges to market regulators. Algorithmic traders require substantial infrastructure and specialised skills to find trading patterns from BIG data as well as automated trading software to exploit high frequency patterns. On the other hand, the “flash crash” in 2010 has urged market regulators to evaluate how an algorithmic trading strategy is going to impact market quality (efficiency and fairness) before it is adopted. Alluvial provides an integrated solution for both user groups.

The Alluvial system is client-server software with web accessibility. Advanced users, who require direct connectivity to their market data feed and dedicated line for order submission, can have the software deployed onsite. A web portal may also be provided for users to access the Backtester, Compliance manager and inventory manager remotely. For general users who may only wish to access the Alluvial Backtester, a lite version of Alluvial which operates as a hosted service is currently under being developed.

Uptick - 4GL Language for Market Data Analysis

In order to enable non-IT users to develop their own back-testable trading algorithms and market quality metrics, the CMCRC developed the UPTICK language, which is the core component of the Alluvial Back-testing Platform. UPTICK is a python based 4GL language. It uses more readable and simplified syntax and has a large number of common financial market concepts and trading attributes (such as true price, spread, VWAP) as inbuilt functions. Users will be able to submit metrics and algorithms developed in UPTICK to the CMCRC infrastructure for processing and visualisation.