Sylvain Benoit

Assistant Professor, Université Paris - Dauphine

Applied Time Series

Les objectifs du cours

The objective of the course is to study the theory, modeling, programming, and interpretation of the major time series models. Some applications to finance will be undertaken using Python. At the end of this class, students should be able to

  • Develop knowledge of basic univariate time series models appropriate for economic and financial data.
  • Learn how to specify and estimate a time series model on these data.
  • Be able to use such models for forecasting and to evaluate their performance.
  • Familiarize with common volatility modelling approaches.

Prerequisites

Students must be enrolled in course Python Programming and must have past Introduction to Financial Econometrics
 

Plan du cours

1/ Times series building blocks.

  • Stationarity
  • Autocorrelation and white noise
  • Testing autocorrelation
  • Non-stationarity

2/ ARMA Framework.

  • Moving average process
  • Auto regressive process
  • ARMA models and the Box-Jenkins method
  • Maximum likelihood estimation
  • Application: Small stocks return

3/ Specific topics and applications

  • Unit-roots
  • Trends
  • Seasonnality

4/ Volatility models

  • GARCH
  • Stochastic volatility (Stock index volatility)
  • MS Switching
     

Bibliographie

Brooks C (2008), Introductory econometrics for Finance, Cambridge Univ Pr.
Brockwell, P.J. and Davis, R.A. (2002), Introduction to time series and forecasting, Springer Verlag.
Campbell J., Lo A., McKinley, A. (1997), The Econometrics of Financial Markets. NJ: Princeton University Press.
Francq C, Zakoïan J.M. (2010), Garch models: Structure, statistical inference and financial applications, Wiley.
Hamilton J. D. (1994), Time Series Analysis, Princeton University Press.
 

Examen

Assignment (40%) + Final Exam (60%)