About CREATES

Center for Research in Econometric Analysis of Time Series, CREATES, is a research unit at Aarhus BSS, hosted by the Department of Economics and Business Economics. The center is funded as a Center of Excellence by the Danish National Research Foundation via a 80 mio DKK (10.74 mio Euros) grant plus own financing. The funding period covers a 10-year period starting 1 April, 2007.

The core group of members includes time series and financial econometricians from Aarhus University (Department of Economics and Business Economics) and the University of Copenhagen (Department of Mathematical Statistics).

CREATES will bring together some of the world’s leading experts in time series and financial econometrics. Over the years, Aarhus University and University of Copenhagen have produced econometrics candidates who today are amongst the most cited researchers in the profession and with remarkable academic careers. Also, within the past few years promising Ph.D. candidates in econometrics have been produced, who are now affiliated with some of the best universities internationally. CREATES brings together these researchers. In addition, a number of other international world leading researchers are affiliated with CREATES as research fellows or associates. Typical for all participating researchers is that there are close ties linking them together: Researchers have already close links to CREATES either via their educational background and/or via research co-operation with other CREATES members.

A major feature of CREATES is the mix of junior and experienced senior researchers. It is pivotal on CREATES’ agenda to put significant efforts into high-level graduate (PhD) training and to support young promising researchers via post doctoral scholarships.

Basic Research Pillars

In macroeconomics and financial economics, as well as in many other disciplines in economics, empirical research is largely based on time series data. Since the work of 1989 Economics Nobel Laureate Trygve Haavelmo, it has been commonplace to view economic time series as realizations of stochastic processes. This implies that it is possible to use statistical inference in constructing and testing models describing the relationship between economic variables. Econometrics is frequently defined as the discipline within the economics science where mathematical and statistical methods are used to quantify and test economic theories. In particular, time series econometrics is concerned with such analysis when the underlying data are based on observations of economic variables measured over a fairly long period of time, e.g. exchange rates, interest rates, stock prices, unemployment figures, and aggregate growth data. This is in contrast to microeconometrics which is largely based on statistical analysis of a large amount of variables for e.g. individuals and firms at a single or a few points in time.

Over the past 25 years, the field of time series econometrics has developed tremendously. It is probably the research discipline within the econometrics profession which has been most influential in terms of both theoretical and practical impact. This development is mainly due to two major advances: The development of a theory to deal with non-stationary data and the development of a theory to model time varying volatility. Both developments have had vast impact, particularly in empirical macroeconomics and finance. In 2003 Prof. Clive W. J. Granger and Prof. Robert F. Engle were awarded the Nobel Prize in Economics for their contributions to these fields. The research to be undertaken within CREATES covers a broad range of topics based on the research pillars time series and financial econometrics, and relate in particular to extensions and further developments of the areas mentioned above.

The focused research areas can be categorized as follows:

  • Theoretical time series econometrics with particular emphasis on long memory, persistence, and optimal inference
  • Empirical modelling of asset returns and volatility
  • Non-linear time series modelling
  • Time Series Forecasting.