The goal of this course is twofold first to enlighten the foundation for econometrics, probability. theory and statistical inference and second to introduce practical econometrics.
This course contains a) the first two weeks of a brief overview of probability theory and statistical inference b), the 6 weeks of classical econometrics with emphasis on time series econometrics.
Students are expected to do computer exercises (RATS MATLAB EVIEWS or any software that you prefer) and written exercises. There will be a 3 hour written exam in part I. The exam accounts for 70% of the overall evaluation for the course the remaining 30% will be based on the exercises.
Paul Ruud: Classical Econometric Theory, Oxford University Press, 2000.
The teacher may distribute some additional material
Some other useful texts are:
For the overview of statistical theory books on probability, statistical inference or general statiststics
will do. For probability theory see, e.g.
Grimmet, G.,R., and Stirzaker, D., R.: Probability and Random Processes
For statistical inference see, e.g.
Casella, G. and Berger, R.,L.: Statistical Inference
Other books on probability or statistics may be useful such as books with time series in the title.
Hamilton, J.D.: Time Series Analysis, Princeton University Press, 1994.
Harvey, A.C.: Time Series Models, Harvester Wheatsleaf, 1993.
Brockwell, P.J. and Davis, R.A.: Time Series: Theory and Methods, Springer 1991.
Poirier, D.J.: Intermediate Statistics and Econometrics, MIT Press 1995.
Enders, W.: Applied Econometric Time Series, John Wiley and Sons, 1995.
Week 3-4: Statistical inference, likelihood, estimation, point estimates, confidence intervals, testing hypotheses, asymptotics etc.
Week 5: Introduction to linear models
Week 6: Generalization of linear models
Week 7: Introducing time series, stationarity, time domain, ARMA and ARIMA.
Week 8: Estimation and testing in ARMA models.
Week 9: Modelling economic time series.
Week 10. Exogeneity, non-stationary models, cointegration.
Sample programs for ARIMA analysis and ARCH
To run the Johansen co-integration regression, use CATS in RATS. Unfortunately
I have not been able to find the new version on our network. An example
of how to use the old code is here
A matlab(octave/scilab) code that generates random walk is here .
A matlab(octave/scilab) starter-code to do Johansen cointegration is here .