Algorithmic Trading


In the 21st century computers are transforming the market place. The dominant factors in trading are computer science, mathematics and statistics. Brokers standing on the market floor shouting under pressure are being replaced by computerized high frequency automatic trading algorithms.

High frequency trading is done by sophisticated trading algorithms that operate 24/7 electronically. They mostly do their work with no human intervention, which can have significant effects on the markets, such as increase their volatility. They conduct as much as 70% of all global trade. They can also cause crashes and extreme price changes. They usually run on servers with low latency and connect to multiple stock and exchange markets via their application protocol interfaces and they make decisions in a split of a second on trading.

Primary aims

The primary aim of this project is to study high frequency trading algorithms in order to improve them in various aspects. Today, most known algorithms combine elements of artificial intelligence, machine learning, natural language processing and data mining. The algorithms usually analyze past market data, social media events and other factors in near real time to find patterns and trends using an auto-regressive learning model with gradient descent and thus they attempt to predict market movements. Their strategies are tested with various configuration on high performance grid computers and the results are fed back in order to improve their configuration. This way it is possible to train the algorithms and evolve them as they gather and analyse more and more data.
Some of the existing trading strategies include prediction of markets, linear regression, game-theory, pettern recognition, neural networks, genetic programming. High frequency trading algorithms include Market making, Ticker Tape trading, Event Arbitrage, Layering, High-frequency stastitical arbitrage.

Secondary aims

The research might involve writing and experimenting with a simple bot for testing purposes. The bot would gather and save data to a database, scheduled to analyze it, and attempt to adjust its configuration. The bot might rely on Bitcoin to conduct its virtual trading. Bitcoin is new open source Internet currency. Bitcoin exchanges function digitally with no down time. Transactions are carried out distributed, with no central supervision, at almost no cost, wholly anonymously and near instantly. This is a very interesting environment to study, unlike any other market places and may have the potential to revolutionise algorithmic trading.


The research attempts to answer the following questions.
- What features should the algorithm have. What data sources should it take into account and how should it behave on that data. What should the decision process be.
- How accurate is the algorithm. What is the optimal configuration. How to measure performance.
- How to correlate social media events with market movements. How to predict volatility, price and volume of trade across multiple markets.
- How to make the algorithm fast enough to adapt to changing markets and social media events in real time.
- How to process large quantities of data effectively. I.e. from tweets, blogs, news, media sources as well as financial data streams from multiple markets.


The research will build on the following skills.
1. Experience in artificial intelligence and machine learning algorithms
2. Experience in natural language processing
3. Experience in data mining and financial market analysis
4. Experience in writing large scale database driven server side applications.

Further information
Role of trading algorithms in computer science and finance