Deriving Dynamic Marketing Effectiveness from Econometric Time Series Models

To understand the relevance of marketing efforts, it has become standard practice to estimatethe long-run and short-run effects of the marketing-mix, using, say, weekly scanner data. Acommon vehicle for this purpose is an econometric time series model. Issues that areaddressed in the literature are unit roots, cointegration, structural breaks and impulse responsefunctions. In this paper we summarize the most important concepts by reviewing all possibleempirical cases that can be encountered in practice using a prototypical model. We provideguidelines for practitioners, and illustrate these for a detailed workedout example