Duration: 3 hours for 13 weeks [ECTS: 6]

Course Description:

The ongoing 4th Industrial Revolution generates and utilizes extensive amounts of data, the so-called Big Data. Big Data Analysis is becoming more and more relevant to financial markets worldwide. It is often said that data are the “new gold” or more accurately the “new oil” for modern digitized economies. The Big Data Analytics course focuses on practical methods to acquire financial data, to process them effectively and visualize the outcome of the analysis. The course begins with an introduction to Big Data and discuss their significance in financial decision-making. A brief presentation of the basic principles of Data Science follows. Simple examples with the Python programming language help students to familiarize themselves with the algorithmic approach of the course. Anaconda, the popular platform for Data Analysis is demonstrated as it will be used to program financial algorithms. The most important libraries for Data Analytics are thoroughly explained, emphasizing the Pandas, NumPy, SciPy and Matplotlib Python packages with various examples. The necessary financial data of historical stock transactions are integrated to the course’s case-studies using the Yahoo! Finance and the Quandl datasets. Students are expected to become proficient at algorithmic trading strategies development using the Zipline platform. The algorithmic trading strategies are evaluated using backtesting over historical data. The results are assessed according to various economic indicators to determine the best trading strategy. Finally, the outcomes of the simulations are appraised using the PyFolio package.

Course outline:

• The implications of the 4th industrial revolution in the financial sector
• Introduction to Data Science
• The Anaconda platform for scientific computing and data analysis
• Quick revision of Python programming
• Financial data repositories and stocks market historical data
• Python packages for scientific computing and data analysis (Pandas, NumPi, SciPy, Matplotlib)
• Testing trading strategies using the Zipline platform
• Evaluation the performance of portfolio diversification using backtesting
• Financial data analysis and visualization with the PyFolio python package

Learning outcomes:

The students upon the completion of the course will be able to:

• Use the Python programming language for data analysis.
• Acquire financial data from online and offline datasets.
• Analyze financial datasets using data science techniques.
• Implement in Python algorithmic trading strategies.
• Assess the effectiveness of algorithmic trading strategies using backtesting.