The Components of Time Series (Course code 0431)


Most of the data published by statistical agencies consist of time series, which is of figures measuring the evolution of socio-economic variables through time. In modern economies, time series data assist governments, businesses and socio-economic actors in their decision making. Based on the movements recorded in time series, governments initiate policies designed to curb unemployment, inflation, etc.; corporations accelerate or slow down the production of goods and services; unions closely monitor the labour market situation to negotiate appropriate wages. Even consumers, more or less systematically, use time series to decide whether the time is right to purchase a house, an automobile; whether to look for a job, etc. Thus, a good understanding of time series translate into better decision-making by everyone and into increased prosperity.

Benefits to participants

This course will enable the participants to recognize, understand and interpret the movements present in time series; and familiarize them with the graphical representation of data.

Target population

The course targets a broad audience: professional and semi-professional social scientists and statisticians, authors and editors of publications. The content of the course is relatively non-technical but provides notions critical to the understanding of time series.

Course outline

The course examines in depth the components of time series:

  • the trend, which reflects the long term evolution of the variable of interest,
  • the business cycle, which reflects current conditions, e.g. prosperity, recession,
  • seasonality, which originates from climatic and institutional factors and tends to recur year after year in a predictable manner,
  • the trading-day variations caused by the different relative importance of days of the week, and other calendar variations caused by changes in the dates of holidays, e.g. Easter.

The course also stresses the meaning and limitations of same-month (from year to year) and of month-to-month comparisons, in the presence of seasonality and other series components. This course assumes that the components of time series are known and does not cover the estimation of the components. That is done in a more technical and specialized course on seasonal adjustment.

Other Related Courses

The course is a desirable prerequisite for other courses, namely STC0434 Seasonal Adjustment with the X-11-ARIMA Method and STC0433 ARIMA Modelling and Forecasting of Time Series.

Duration: 1 ½ days

Date modified: