Stock Trading

The Client is an artificial intelligence (AI) platform that matches corporate executives with the institutional investors that are most likely to buy, or sell, their stock in the next 90 days — while simultaneously offering all users data discovery technologies to efficiently prepare for all types of corporate-investor access events.

The Client was seeking an intelligent Financial Prediction Model for extracting investor behavior in the USA stock market, in order to analyze institutional investors’ interests.

What services did Data-Core provide?

The key objective was to develop an intelligent Financial Prediction Model for extracting investor behavior in the USA stock market, so as to increase the availability of decision support data and hence increase investor satisfaction. The model was built based on quarterly data and deals with stock portfolio analysis of the institutional investors.

The Financial Prediction Model can:

• Individually evaluate the relationship with an investor and a stock for Owner & Non-Owner models in the Russell 3000
• Use Azure data mining model and statistical technique (Algorithm)
• Handle “imbalance classification of data using R-scripts” in MS Azure ML Studio

  The Challenge was to identify characteristics of investors that are most predictive in signaling that they are likely incremental buyers, as well as those characteristics that suggest an investor is at risk of reducing their position.

• Breadth: Be able to get 30 recommendations for each company in the Russell 3000 (10 new buys / 10 incremental buys / 10 incremental sells)
• Accuracy: Move accuracy from current ~54% to >65%
• Automation of scripts for data preparation

In Summary…

Data-Core provided the Client interactive & integrated dashboards to help investors and corporates more efficiently prepare for meetings by quickly identifying themes, trends and outlier data that is most likely to warrant discussion. This was accomplished with easy to use analytics to quickly compare fundamental, valuation and sentiment factors relative to peers. Data-Core’s simple but meaningful designing of dashboards was a huge satisfaction to the Client which then got promoted to one of the most complicated Dundas BI dashboards in the industry, followed by data migration from MySQL to Data Warehouse. Data-Core continues to provide maintenance to the developed dashboards.