Stochastic Processes and Their Applications in Economics and the Social Sciences

Overview

The aim of this module is to give an introduction into the statistical methods of the social and economic sciences, with special emphasis on the analysis of time series data..

By the end of this module, a student should understand

• which methods are used to collect and analyse time series data.

Syllabus and time schedule of the lecture course (the time schedule will be completed according to the progress of the course)

1. Time series: Overview (08-15/11/2006)
2. Time series: Component Methods (Local and Global) (15/11/2006, discussion class: 05/12/2006)
3. Stochastic Processes: Overview (22/11/2006)
4. Stochastic Processes: White Noise, Random Walk, General Linear Processes
5. The Concept of Stationarity
6. Time and Frequency
7. The ARIMA Model
8. Transformation of Time Series and Stochastic Processes, Filters
9. Prediction (20/12/2006, 10/01/2007)
10. Markov Processes (10-17/01/2007) ** no lecture on 31/01/2007 because of eGov Day 2007
11. Stochastic Processes in Continuous Time and Space (24/01/2007)
12. Stochastic Processes in Queueing Theory (07/02/2007)
13. Oral examination (22/02/2007, 15:00-17:20, no lecture on 14/02/2007)

Course material

All slides for this course can be found here (chapters 1 through 9) and here (chapters 10 through 12).

Basic Reading

• Reiner Schlittgen, Bernd H.J. Streitberg (2001): Zeitreihenanalyse, München: Oldenbourg 9. Aufl. 2001
• Peter J. Brockwell, Richard A. Davis (1991): Time Series: Theory and Methods, New York: Springer, 2nd edition
• Peter J. Brockwell, Richard A. Davis (2002): Introduction to Time Series and Forecasting, New York: Springer, 2nd edition

SPSS

The data analysis software system SPSS is available in all labs using XP:

All Programs -> SPSS for Windows --> SPSS 14.0 for Windows

Additional information on SPSS:

• SPSS Homepage
• SPSS page (University of Cologne)

Data Sets

Several data sets will be used in the tutorial

• Electricity production in Germany 1955-1979
  This time series can be analysed with the help of the component method.
• Stock indexes
  10 different time series describing the most important stock indexes collected from november 1995 to February 2001
  • Dow Jones Industrial
  • S&P Composite
  • Nikkei 30
  • Dax 30
  • CAC 40
  • Swiss Market
  • Milan MIB 30
  • Ibex 35I
  • BEL 20
  • FTSE 100

  Data set

• others

Time schedule for the labs (to be completed later on)