Capital Alliance (Pvt) Limited (CAL), is one of the leading investment institute in Sri Lanka active in primary as well as equity markets, with a strategic interest in using financial data analytics to improve the predictions of financial markets thus improving both quality and return-on-investment. As one of the leading higher education institute in Sri Lanka, SLIIT has always been extremely keen on building an enriching academic and research, and committed to fostering and promoting new technologies such as Data Science and Machinery Learning. Therefore, SLIIT and CAL are collaborating as willing partners with a common goal; to initiate a Financial Data Analytics research lab, allowing those interested to conduct ample and extensive research of both academic and industrial relevance. Financial data analytics is mainly focused on the application of machine-learning and statistical techniques to learn from past financial data, in order to make useful predictions about key financial indicators.

The knowledge gained from such predictions may then be applied to make prudent investment decisions. Although financial data analytics is a prominent sub-field of data analytics, it is rarely utilised as a tool in the primary and secondary financial markets in Sri Lanka. Hence, the objective of setting up this joint research entity is to explore the potential directions current and future research can take, and to build local expertise in the area of financial data analytics in order to make better investment decisions. Furthermore, the knowledge assimilated and created throughout the process may be shared with the academic community via reputable scientific journals, effectively improving the visibility of the local research conducted within this domain. During the first phase of the joint research, there will be two main focused areas. They are making financial transactions based on macro-economic indicators and news events.

Macro-economic indicators such as inflation and employment rate can be difficult to predict using historical data. These factors are the result of highly complex interactions between multiple forces that affect an economy. Likewise, the behaviour of financial markets such as equity markets and forex markets may be highly volatile, with multiple factors affecting them. This would make it extremely difficult to predict their behaviour by identifying the trends and patterns of the past data alone. In this project, we try to create a learning model to capture the effect on the ‘sentiment’ of macroeconomic indicators and financial markets, through public domain information sources such as financial news reports and opinion pieces and try to incorporate those sentiment values into improving the accuracy of the prediction models. News events provide a critical information source in predicting the behaviour of financial markets. Time series prediction models have been used to predict the variations in these financial markets.

Some of these models do capture external news events to improve the accuracy of such predictions. However, it is not entirely accurate to consider the impact of news events at the same level as historical financial market data, such as share prices or macro-economic indicators like inflation rate. The reason for this, is that a data point in a financial market is captured and considered at a snapshot in time, while news event may affect multiple data points over a longer period of time. Moreover, the effect of these news event may have a time-diminishing effect. In this project, we try to capture this time-diminishing effect of news events at a more fundamental level than the historical data points, to improve the accuracy of predictions done on financial markets. The location proposed for the research lab will be provided by SLIIT and led by key personnel identified by SLIIT, comprising both internal staff members as well as external experts.

Additionally, the lab will employ fulltime research engineers and MPhil/PhD candidates who will engage in cutting-edge research in the field of financial data analytics. As one of the first of its kind in the island, the lab is undoubtedly a significant milestone all-round, and paves the way for a more academically sound approach in financial data analytics. Bolstering the institute’s academic presence and enabling them uphold their promise of educational excellence, SLIIT together with CAL, has proudly launched the Financial Intelligent Research joint-lab. The agreement which was signed at Cinnamon Lakeside by Prof. Lalith Gamage, Vice Chancellor/ CEO, SLIIT and Mr. Ajith, Fernando, the Chairman of Capital Alliance Limited, is a significant step towards development and enhanced support for students so they may be equipped with the best possible tools within this field. Both SLIIT and CAL strongly believe that this partnership will nurture this stream of study in a unique and innovative manner, and look forward to seeing students pursuing this field thrive under it.

The joint research lab will be led by a team of senior academics of SLIIT including Prof. Lalith Gamage, Prof. Chandimal Jayawardena, and Dr. Dharshana Kasthurirathna. Prof. Lalith Gamage is the Vice Chancellor/CEO and founder of SLIIT, and has over 35 years of research and leadership experience in the areas of artificial intelligence, machine-learning, and robotics. Prof. Chandimal Jayawardena is the Dean of the Faculty of Computing, and bears 20 years of research & industry experience in the areas of robotics, intelligent systems and machine-learning. Dr. Darshana Kasthurirathna is a senior lecturer of the Department of Software Engineering, and has 14 years of experience as a software engineer and researcher. His areas of expertise are complex systems and data analytics.