Course description

Financial econometrics is the intersection of statistical techniques and finance. Financial econometrics provides a set of tools that are useful for modelling financial data and testing beliefs about how markets work and prices are formed. Conversely, new techniques in analyzing financial data can lead to empirical facts inconsistent with existing theories, begging for new models or investment strategies.

The course begins with models of time varying expected returns which are useful in formulating expected returns. A casual follower of financial asset prices quickly notices prolonged periods of high volatility followed by more tranquil periods. We next develop several modelling tools that allow us to forecast or predict risk, or volatility, when risk is changing through time. The financial crisis highlighted contagion or the fact that returns on assets tend to be more highly correlated in market downturns. Some market prices must satisfy long run relationships.

Who should take this class?

This course will be useful to students who plan to take empirically oriented finance courses as well as students who want to get a solid understanding of the tools required to analyze and model financial asset prices. The link between new statistical models and implementation is emphasized throughout. The course is also recommended for students who will implement empirical work in their theses, since the material will cover most of the necessary statistical tools required for a successful writing.

Textbook or notes

Lectures are based on lecture notes.

Software

The students will be using either software available in our lab, while the R language/software will be shown through applications.

Attendance

Attendance of the class is required and it is essential.  The course materials are mainly from the notes. Many conceptual issues and financial econometrics thinking are only taught in the class.

Topics covered

1. Predicting financial returns
2. Regression models
3. Autoregressive models
4. Moving average models
5. ARMA
6. Model evaluation
7. Predicting return volatility risk
8. ARCH
9. ARCH-in-mean
10. GARCH
11. EGARCH and other variations
12. Review of VAR/VECM models, Granger causality, multivariate GARCH.
13. Predicting risk and returns for multiple assets
14. Vector models for the mean
15. Random walks and cointegration
16. Cointegration and error correction models
17. Forecasting cointegrated systems
18. Linear time series and dynamics of returns
19. Efficient portolios and CAPM
20. Consumption-based CAPM
21. Testing asset pricing models: GMM estimates
22. Introduction to non-linear econometric models
23. Bilinear models, piecewise linear models, TAR, STAR, SETAR and their applications.
24. Panel data
25. Special topics
26. Seasonality
27. SVAR models
28. VAR modeling: Impulse responses, variance decompositions
29. Forecasting criteria

e. Logit and Probit models