Günter Lang
Werbemarkt Fernsehen:
Zur Eignung der Spektralanalyse als Prognoseinstrument
Abstract:
Over more than a decade, advertising rates per 1000 viewers, television
consumption as well as the number of advertising spots have been
steadily increasing. As a consequence, television has developed to the
most important medium for the advertising industry and attracts a 40%
share of German gross advertising spending. Motivated by the recent
slump of advertising rates and of the number of spots, this paper
attempts to develop a forecast model for real advertising spending on
the German TV market. In a first step, spectral analysis is used to
identify the most important cycles on the advertising market. In a
second step, the identified cycles are entering a regression model
which is the basis for making forecasts. To evaluate the forecast
quality, the results are compared to a standard ARIMA model. The
estimations are based on monthly data of the German TV advertising
market from 1990 to 2004. Actually, the results show that the estimated
cyclical pattern describes the trends on the TV advertising market very
well. The cycle model is therefore a useful tool for making ex-ante
forecasts of real advertising spending. Underlining the quality of the
approach, the ARIMA approach is performing significantly worse than the
introduced cycle model.
JEL: L82, M37, E32
Paper:
Paper available as pdf-file.
Beitrag Nr. 274, Volkswirtschaftliche Diskussionsreihe, Institut
für
Volkswirtschaftslehre der Universität Augsburg
Contact:
Günter
Lang, University of Augsburg, Department of Economics, D-86135
Augsburg,
Germany, phone +49-821-598-4195, fax +49-821-598-4230,
email: guenter.lang@wiwi.uni-augsburg.de
v.
K., 12.03.2005