• Home
• Contact
• People
• News
• Events
• Services
• Researcher Development Initiative
• Clinics
  • Materials
  • Data
  • Dependencies
  • Impact
  • Measurement
  • Scaling
  • Questions
  • Regression
  • Segments
  • Text
  • Time
• Postgraduate Study

Time: Time Series Analysis

Instructors

• Prof. Cees van der Eijk (Methods and Data Institute)

Content

To examine the ways in which series of regularly repeated observations can be analysed (e.g. statistics on economic indicators, crime, transportation, government support, etc.). In doing so, the clinic will explore the kinds of questions addressed by (quasi-)time-series analysis, such as:

• is there any kind of systematic pattern in the fluctuations over time, for example trend or cyclical pattern?
• is the variation in a series dependent on that in another one? For example, is support for the government driven by economic developments?
• does it respond to external events and if so, how quickly and how strongly? For example, does consumption, investment and production react to changes in taxes?

The term time-series is often used for large numbers of repeated observations (at least 50), while ‘quasi-time series’ contain smaller numbers of observations which generate somewhat different problems and necessitate somewhat different procedures.