Methodology for Identifying Time-Trend Patterns; Application to Advertising Expenditure of 28 European Countries in the Period 1994-2004  

Katarina Košmelj
Biotechnical Faculty, University of Ljubljana, Ljubljana, Slovenia

Vesna Žabkar
Faculty of Economics,University of Ljubljana, Ljubljana, Slovenia

The aim of our study is to reveal different time-trend patterns in the ratio of advertising expenditure over gross domestic product for 28 European countries in the period 1994-2004.

To fulfil the objective we used two multidimensional statistical approaches: cluster analysis and multidimensional scaling of time-varying data, followed by a linear regression analysis of the time-series within the clusters. The proximity matrix was calculated using the dissimilarity between two time series, which takes into account the order of time points, the information on the fixed time points, and the weights which are arbitrary. We identified four clusters of countries with similar trend patterns in 1994-2004: awakening countries, stable countries, catching-up countries and the leading cluster.