MSc in Accounting and Finance, MSc in Accounting Taxation and Financial Management, Department of Accounting and Finance
In the first part of the course, a parametric approach to financial models is introduced. It covers equilibrium and dynamic models while analyzing various estimation methods. The predictive capability of parametric models is assessed, with a focus on the relationship between short-term and long-term forecasting horizons. Laboratory applications are primarily conducted using EVIEWS software.
The second part of the course explores the use of neural networks, the most advanced non-parametric approach, in financial applications, with an emphasis on forecasting techniques. The following topics are covered in detail:
- Neural networks and financial innovation
- Fundamental concepts and basic neural network models
- Backpropagation neural networks
- Non-parametric estimation using neural networks
- Estimation process of a neural network model
- Confidence and prediction intervals
- Estimation of conditional means
- Neural networks as an extension of multifactor models (APT and Multifactor CAPM)
- Volatilities and correlations
- Data preprocessing
- Time series forecasting and classification applications
- Model evaluation and sensitivity analysis
- Current technologies and systems
Alongside the theoretical analysis, specialized software BPSim and MATLAB is used in a range of financial applications, including corporate financial management, securities trading, investment management, and financial risk management.