DGPE PhD course: Change-point and mixture Models with applications to macroeconomic time series and option pricing

29 October 2014. Lecturer: Associate Professor Jeroen V.K. Rombouts, ESSEC Business School

2014.01.03 | Solveig Nygaard Sørensen

Date Wed 29 Oct
Time 09:00 16:00
Location Fuglesangs Allé 4, 8210 Aarhus V, bldg. 2610, room S313B

It is well known that econometric models with a fixed structural form or constant parameters are likely miss-specified when estimated on long samples. This has severe consequences for the interpretation of the models and their forecasting performance. For example, the estimates of a GARCH volatility model may suffer from a substantial upward bias in the persistence parameters because the form of the conditional variance is relatively inflexible and held fixed throughout the entire sample period.  

Finite mixture, Markov Switching and change-point models are flexible parametric models that can capture many time series features. We will cover several models used for financial and macroeconomic time series. Inference for the models will be classical or Bayesian depending on the complexity of the likelihood. However no prior knowledge of Bayesian techniques is required. The applications we will involve macroeconomic time series and option prices.

Registration

  • Registration via webshop (closed)

The course is free of charge for DGPE members from AU, KU, CBS, SDU, and Nordic universities. The course fee is EUR 200 for other participants.

Contact

Lecturer: Associate Professor Jeroen V.K. Rombouts, ESSEC Business School
Email: rombouts@essec.edu

Administrative support: Solveig Nygaard Sørensen, sns@creates.au.dk

DGPE Courses