Analysis of financial data is an exciting application area for data science. You will study linear, non-linear and density models of sequential data, using neural networks.
You will use gradient descent techniques and apply Kalman filters to enable proper dynamic treatments. You will investigate the concepts of overfitting, generalisation and performance evaluation. You will see several practical applications of neural networks and extended Kalman filters from the FinTech area of financial data analytics. These include value-at-risk estimation, option pricing, portfolio estimation and automated, algorithmic trading.
- Introduction to Financial Data Modelling
- Financial time series models
- Learning algorithms for financial data models
- Density modelling with financial data
- Performance estimation of nonlinear models
- Extended Kalman filtering of nonlinear models
- Value-at-Risk estimation
- Portfolio selection and estimation
- Automated algorithmic trading
- Introduction to FinTech
15 (150 hours)
- Summative coursework (30%)
- Written examination (70%)