Options Risk Management
Large multinational banks often use a variety of instruments in order to protect themselves from risk. One of the most common ones is options. These are contracts which allow the buyer the right but not the obligation to exchange the stock at a specified price on a specific date. These are used typically to manage risk and ensure that the bank is not overexposed to market risk in any specific sector.

Detecting Market State
The type of tool used for hedging risk is dependent primarily on the state of the market. The client we were working with wanted to use reinforcement learning – the same type of learning that was used in game engines such as Alpha Go. They wanted an algorithm developed that could detect the “state” of the market and prompt traders to act accordingly based on historical data.

Variable Selection
A large number of variables were taken into consideration for this problem such as the stock price, momentum, earnings releases and 10-K reports, this data was collated and aggregated into a standardised format. At the industry level, these companies were grouped and statistics were calculated such as the average volatility of a stock in that industry. Integration was then done with a live Bloomberg price feed to give traders an indication of the state of the market – bullish or bearish for a given stock based on this information.

Solution
An automated hybrid solution built on top of Python was created to allow the client to anticipate unusual movements of the market would allow them to better provide their services and manage the risk associated with it. Interactive graphical tools were built in Java Swing which integrated with these Python services in order for the floor traders to better assess the best course of action to take in a specific scenario.

Our Results
DataSpartan delivered both research results for this case and an interactive visualisation of maker movements to their requirements. This tool allows them to better anticipate shock and explain tendencies. The current tool is integrated with on the dashboards of options traders as a customisation widget – documentation was also provided to the client in-house development team in case they wanted to expand it further.

Search
Since the advent of Google, search has been the bedrock for the spur of the information age. With the touch of a button it is possible to access any public indexed document stored. The drawback with conventional search engines such as Google and Bing is that they cannot be used on proprietary data sets.

Compliance
In the context of compliance, financial institutions usually have a compliance team monitor and overlook other teams to ensure that they are not in breach of any regulations. In the case of the client, they wanted to save time by ensuring that their compliance team was able to find the right documentation. The client had an internal database of PDF files which contained information about whether or not a statement or comment should be flagged.

Phrases
There are certain keywords and phrases that traders would use that corresponded to a specific set of stocks. This would amount to insider trading and the job of the compliance team was to closely monitor these Bloomberg chat logs to ensure that this was not the case. There were a very high number of false flags which meant that the team had to spend time trying to identify the relevant document to check if it was a real flag.

Solution
DataSpartan was asked to solve the search part of the problem by creating a tool which would save the compliance officers time by allowing them to search their internal databases more quickly for the relevant information. A custom interface was built using Django which had PDF previews for key words and phrases to allow the officers to preview the documents manually and by eye for relevancy.

Our Result
This is the first component of a larger system which involves a document recommendation engine highlighting the usage of the keywords in the document explicitly. The component being developed is currently being integrated into the client servers and is projected to save each officer 3 hours of work each month. Full integration of the larger system is projected to save 10 hours of work each month.

Client Classification
For business to business enterprises, client classification focuses on putting a customer into a specific category or bucket. In this specific scenario, DataSpartan was working with a crowdsourcing startup who wanted to identify whether or not a business on their platform was likely to “hit” their crowdfunded target amount raised. The client also wanted us to inform them of which variables were most important in terms of increasing the likelihood of a raise.

Problem
A large proportion of the onboarding process was done manually and it was extremely time consuming for the startup to fully onboard each client only to have them not complete a successful raise. Standardized data was collected about the size of the company, the number of followers they had in social media as well as financial performance, their team, industry and competition as well as exit potential. Currently, 56% of the companies that were going through the funnel had a successful raise.

Solution
The first part of the problem was identifying which variables were the most likely to influence the outcome. Historical data was extracted, cleaned and standardised to allow this analysis and three variables were singled out which determined the probability of a successful raise. A supervised learning model was then created which could classify these leads based on their likelihood of conversion based on their data. Python was used for this portion of the data analysis and libraries such as NumPy and SciPy were leveraged to ensure rapid computation and fast model prototyping.

Our Results
The current research and results have been incorporated back into their lead pipeline in the form of a qualifying questionnaire which filters out unlikely leads and pre categorises the clients. This has resulted in a 4% decrease in the number of man hours used as well as a 6% increase in the number of companies that successfully raise. Further work is being done to automate some of the manual processes such as KYC (Know-Your-Client) document verification to allow the client to efficiently scale up its business. Work is also being done to improve the data pipeline and to ensure that the data collected is being stored in the appropriate format for analysis.

Purpose
Our client lacked the in-house talent required to develop a generic trading platform for back testing strategies. DataSpartan was brought in to resource and manage the project.

Approach
The backtesting system was built in Java and used Spark Streaming, Kafka and HDFS to build a continuous exchange-transform-load (ETL) pipeline for gathering the data feeds of stock prices and in particular streaming social media data in order to produce short term predictions of the stock price.

Our Results
The system has been fully built and currently DataSpartan quants are building on top of the system in order to identify profitable strategies in both the high frequency and medium frequency space for market making. Currently, a market-impact tested strategy of Sharpe ratio 1 has been found and is being put into production.